Search Results

Search found 1614 results on 65 pages for 'emps (expensive managemen'.

Page 11/65 | < Previous Page | 7 8 9 10 11 12 13 14 15 16 17 18  | Next Page >

  • SQL SERVER – SOS_SCHEDULER_YIELD – Wait Type – Day 8 of 28

    - by pinaldave
    This is a very interesting wait type and quite often seen as one of the top wait types. Let us discuss this today. From Book On-Line: Occurs when a task voluntarily yields the scheduler for other tasks to execute. During this wait the task is waiting for its quantum to be renewed. SOS_SCHEDULER_YIELD Explanation: SQL Server has multiple threads, and the basic working methodology for SQL Server is that SQL Server does not let any “runnable” thread to starve. Now let us assume SQL Server OS is very busy running threads on all the scheduler. There are always new threads coming up which are ready to run (in other words, runnable). Thread management of the SQL Server is decided by SQL Server and not the operating system. SQL Server runs on non-preemptive mode most of the time, meaning the threads are co-operative and can let other threads to run from time to time by yielding itself. When any thread yields itself for another thread, it creates this wait. If there are more threads, it clearly indicates that the CPU is under pressure. You can fun the following DMV to see how many runnable task counts there are in your system. SELECT scheduler_id, current_tasks_count, runnable_tasks_count, work_queue_count, pending_disk_io_count FROM sys.dm_os_schedulers WHERE scheduler_id < 255 GO If you notice a two-digit number in runnable_tasks_count continuously for long time (not once in a while), you will know that there is CPU pressure. The two-digit number is usually considered as a bad thing; you can read the description of the above DMV over here. Additionally, there are several other counters (%Processor Time and other processor related counters), through which you can refer to so you can validate CPU pressure along with the method explained above. Reducing SOS_SCHEDULER_YIELD wait: This is the trickiest part of this procedure. As discussed, this particular wait type relates to CPU pressure. Increasing more CPU is the solution in simple terms; however, it is not easy to implement this solution. There are other things that you can consider when this wait type is very high. Here is the query where you can find the most expensive query related to CPU from the cache Note: The query that used lots of resources but is not cached will not be caught here. SELECT SUBSTRING(qt.TEXT, (qs.statement_start_offset/2)+1, ((CASE qs.statement_end_offset WHEN -1 THEN DATALENGTH(qt.TEXT) ELSE qs.statement_end_offset END - qs.statement_start_offset)/2)+1), qs.execution_count, qs.total_logical_reads, qs.last_logical_reads, qs.total_logical_writes, qs.last_logical_writes, qs.total_worker_time, qs.last_worker_time, qs.total_elapsed_time/1000000 total_elapsed_time_in_S, qs.last_elapsed_time/1000000 last_elapsed_time_in_S, qs.last_execution_time, qp.query_plan FROM sys.dm_exec_query_stats qs CROSS APPLY sys.dm_exec_sql_text(qs.sql_handle) qt CROSS APPLY sys.dm_exec_query_plan(qs.plan_handle) qp ORDER BY qs.total_worker_time DESC -- CPU time You can find the most expensive queries that are utilizing lots of CPU (from the cache) and you can tune them accordingly. Moreover, you can find the longest running query and attempt to tune them if there is any processor offending code. Additionally, pay attention to total_worker_time because if that is also consistently higher, then  the CPU under too much pressure. You can also check perfmon counters of compilations as they tend to use good amount of CPU. Index rebuild is also a CPU intensive process but we should consider that main cause for this query because that is indeed needed on high transactions OLTP system utilized to reduce fragmentations. Note: The information presented here is from my experience and there is no way that I claim it to be accurate. I suggest reading Book OnLine for further clarification. All of the discussions of Wait Stats in this blog is generic and varies from system to system. It is recommended that you test this on a development server before implementing it to a production server. Reference: Pinal Dave (http://blog.SQLAuthority.com) Filed under: Pinal Dave, PostADay, SQL, SQL Authority, SQL Query, SQL Scripts, SQL Server, SQL Tips and Tricks, SQL Wait Stats, SQL Wait Types, T SQL, Technology

    Read the article

  • Ask the Readers: Backing Your Files Up – Local Storage versus the Cloud

    - by Asian Angel
    Backing up important files is something that all of us should do on a regular basis, but may not have given as much thought to as we should. This week we would like to know if you use local storage, cloud storage, or a combination of both to back your files up. Photo by camknows. For some people local storage media may be the most convenient and/or affordable way to back up their files. Having those files stored on media under your control can also provide a sense of security and peace of mind. But storing your files locally may also have drawbacks if something happens to your storage media. So how do you know whether the benefits outweigh the disadvantages or not? Here are some possible pros and cons that may affect your decision to use local storage to back up your files: Local Storage Pros You are in control of your data Your files are portable and can go with you when needed if using external or flash drives Files are accessible without an internet connection You can easily add more storage capacity as needed (additional drives, etc.) Cons You need to arrange room for your storage media (if you have multiple externals drives, etc.) Possible hardware failure No access to your files if you forget to bring your storage media with you or it is too bulky to bring along Theft and/or loss of home with all contents due to circumstances like fire If you are someone who is always on the go and needs to travel as lightly as possible, cloud storage may be the perfect way for you to back up and access your files. Perhaps your laptop has a hard-drive failure or gets stolen…unhappy events to be sure, but you will still have a copy of your files available. Perhaps a company wants to make sure their records, files, and other information are backed up off site in case of a major hardware or system failure…expensive and/or frustrating to fix if it happens, but once again there is a nice backup ready to go once things are fixed. As with local storage, here are some possible pros and cons that may influence your choice of cloud storage to back up your files: Cloud Storage Pros No need to carry around flash or bulky external drives All of your files are accessible wherever there is an internet connection No need to deal with local storage media (or its’ upkeep) Your files are still safe if your home is broken into or other unfortunate circumstances occur Cons Your files and data are not 100% under your control Possible hardware failure or loss of files on the part of your cloud storage provider (this could include a disgruntled employee wreaking havoc) No access to your files if you do not have an internet connection The cloud storage provider may eventually shutdown due to financial hardship or other unforeseen circumstances The possibility of your files and data being stolen by hackers due to a security breach on the part of your cloud storage provider You may also prefer to try and cover all of the possibilities by using both local and cloud storage to back up your files. If something happens to one, you always have the other to fall back on. Need access to those files at or away from home? As long as you have access to either your storage media or an internet connection, you are good to go. Maybe you are getting ready to choose a backup solution but are not sure which one would work better for you. Here is your chance to ask your fellow HTG readers which one they would recommend. Got a great backup solution already in place? Then be sure to share it with your fellow readers! How-To Geek Polls require Javascript. Please Click Here to View the Poll. Latest Features How-To Geek ETC The 20 Best How-To Geek Explainer Topics for 2010 How to Disable Caps Lock Key in Windows 7 or Vista How to Use the Avira Rescue CD to Clean Your Infected PC The Complete List of iPad Tips, Tricks, and Tutorials Is Your Desktop Printer More Expensive Than Printing Services? 20 OS X Keyboard Shortcuts You Might Not Know Winter Sunset by a Mountain Stream Wallpaper Add Sleek Style to Your Desktop with the Aston Martin Theme for Windows 7 Awesome WebGL Demo – Flight of the Navigator from Mozilla Sunrise on the Alien Desert Planet Wallpaper Add Falling Snow to Webpages with the Snowfall Extension for Opera [Browser Fun] Automatically Keep Up With the Latest Releases from Mozilla Labs in Firefox 4.0

    Read the article

  • To sample or not to sample...

    - by [email protected]
    Ideally, we would know the exact answer to every question. How many people support presidential candidate A vs. B? How many people suffer from H1N1 in a given state? Does this batch of manufactured widgets have any defective parts? Knowing exact answers is expensive in terms of time and money and, in most cases, is impractical if not impossible. Consider asking every person in a region for their candidate preference, testing every person with flu symptoms for H1N1 (assuming every person reported when they had flu symptoms), or destructively testing widgets to determine if they are "good" (leaving no product to sell). Knowing exact answers, fortunately, isn't necessary or even useful in many situations. Understanding the direction of a trend or statistically significant results may be sufficient to answer the underlying question: who is likely to win the election, have we likely reached a critical threshold for flu, or is this batch of widgets good enough to ship? Statistics help us to answer these questions with a certain degree of confidence. This focuses on how we collect data. In data mining, we focus on the use of data, that is data that has already been collected. In some cases, we may have all the data (all purchases made by all customers), in others the data may have been collected using sampling (voters, their demographics and candidate choice). Building data mining models on all of your data can be expensive in terms of time and hardware resources. Consider a company with 40 million customers. Do we need to mine all 40 million customers to get useful data mining models? The quality of models built on all data may be no better than models built on a relatively small sample. Determining how much is a reasonable amount of data involves experimentation. When starting the model building process on large datasets, it is often more efficient to begin with a small sample, perhaps 1000 - 10,000 cases (records) depending on the algorithm, source data, and hardware. This allows you to see quickly what issues might arise with choice of algorithm, algorithm settings, data quality, and need for further data preparation. Instead of waiting for a model on a large dataset to build only to find that the results don't meet expectations, once you are satisfied with the results on the initial sample, you can  take a larger sample to see if model quality improves, and to get a sense of how the algorithm scales to the particular dataset. If model accuracy or quality continues to improve, consider increasing the sample size. Sampling in data mining is also used to produce a held-aside or test dataset for assessing classification and regression model accuracy. Here, we reserve some of the build data (data that includes known target values) to be used for an honest estimate of model error using data the model has not seen before. This sampling transformation is often called a split because the build data is split into two randomly selected sets, often with 60% of the records being used for model building and 40% for testing. Sampling must be performed with care, as it can adversely affect model quality and usability. Even a truly random sample doesn't guarantee that all values are represented in a given attribute. This is particularly troublesome when the attribute with omitted values is the target. A predictive model that has not seen any examples for a particular target value can never predict that target value! For other attributes, values may consist of a single value (a constant attribute) or all unique values (an identifier attribute), each of which may be excluded during mining. Values from categorical predictor attributes that didn't appear in the training data are not used when testing or scoring datasets. In subsequent posts, we'll talk about three sampling techniques using Oracle Database: simple random sampling without replacement, stratified sampling, and simple random sampling with replacement.

    Read the article

  • Is this simple XOR encrypted communication absolutely secure?

    - by user3123061
    Say Alice have 4GB USB flash memory and Peter also have 4GB USB flash memory. They once meet and save on both of memories two files named alice_to_peter.key (2GB) and peter_to_alice.key (2GB) which is randomly generated bits. Then they never meet again and communicate electronicaly. Alice also maintains variable called alice_pointer and Peter maintains variable called peter_pointer which is both initially set to zero. Then when Alice needs to send message to Peter they do: encrypted_message_to_peter[n] = message_to_peter[n] XOR alice_to_peter.key[alice_pointer + n] Where n i n-th byte of message. Then alice_pointer is attached at begining of the encrypted message and (alice_pointer + encrypted message) is sent to Peter and then alice_pointer is incremented by length of message (and for maximum security can be used part of key erased) Peter receives encrypted_message, reads alice_pointer stored at beginning of message and do this: message_to_peter[n] = encrypted_message_to_peter[n] XOR alice_to_peter.key[alice_pointer + n] And for maximum security after reading of message also erases used part of key. - EDIT: In fact this step with this simple algorithm (without integrity check and authentication) decreases security, see Paulo Ebermann post below. When Peter needs to send message to Alice they do analogical steps with peter_to_alice.key and with peter_pointer. With this trivial schema they can send for next 50 years each day 2GB / (50 * 365) = cca 115kB of encrypted data in both directions. If they need more data to send, they simple use larger memory for keys for example with today 2TB harddiscs (1TB keys) is possible to exchange next 50years 60MB/day ! (thats practicaly lots of data for example with using compression its more than hour of high quality voice communication) It Seems to me there is no way for attacker to read encrypted message without keys even if they have infinitely fast computer. because even with infinitely fast computer with brute force they get ever possible message that can fit to length of message, but this is astronomical amount of messages and attacker dont know which of them is actual message. I am right? Is this communication schema really absolutely secure? And if its secure, has this communication method its own name? (I mean XOR encryption is well-known, but whats name of this concrete practical application with use large memories at both communication sides for keys? I am humbly expecting that this application has been invented someone before me :-) ) Note: If its absolutely secure then its amazing because with today low cost large memories it is practicaly much cheeper way of secure communication than expensive quantum cryptography and with equivalent security! EDIT: I think it will be more and more practical in future with lower a lower cost of memories. It can solve secure communication forever. Today you have no certainty if someone succesfuly atack to existing ciphers one year later and make its often expensive implementations unsecure. In many cases before comunication exist step where communicating sides meets personaly, thats time to generate large keys. I think its perfect for military communication for example for communication with submarines which can have installed harddrive with large keys and military central can have harddrive for each submarine they have. It can be also practical in everyday life for example for control your bank account because when you create your account you meet with bank etc.

    Read the article

  • Nest reinvents smoke detectors. Introduces smart and talking smoke detector that keeps quite when you wave

    - by Gopinath
    Nest, the leading smart thermostat maker has introduced a smart home device today- Nest Protect, a smart, talking smoke & carbon monoxide detector that can quite when you wave your hand. Less annoyances and more intelligence Smoke detectors are around for hundreds of years and playing a major role in providing safety from fire accidents at home. But the technology of these devices is stale and there is no major innovation for the past several years. With the introduction of Nest Protect, the landscape of smoke detectors is all set to change just like how Nest thermostat redefined the industry two years ago. Nest Protect is internet enabled and equipped with motion- and smoke-detection sensors so that when it starts beeping you can silence it by waving hand instead of doing circus feats to turn off the alarm. Everyone who cooks in a home equipped with smoke detector would know how annoying it is to turn off sensitive smoke detectors that goes off control quite often. Apart from addressing the annoyances of regular smoke detector, Nest Protect has talking capabilities. It can alert users with clear & actionable instructions when it detects a danger. Instead of harsh beeps it actually speak to you so you know what is happening. It will tell you what smoke it has detected and in which room it is detected. Multiple Nest Protects installed in a home can communicate with each other. Lets say that there is a smoke in bed room, the Nest Protect installed in bed room shares this information to all Nest Protects installed in the home and your kitchen device can alert you that there is a smoke in bed room. There is an App for that The internet enabled Nest Protect has an app to view its status and various alerts. When the Protect is running on low battery it alerts you to replace them soon. If there is a smoke at home and you are away, you will get message alerts. The app works on all major smartphones as well as tablets. Auto shuts down gas furnaces/heaters on smoke Apart from forming a network with other Nest Protect devices installed at home, they can also communicate with Nest Thermostat if it is installed. When carbon monoxide is detected it can shut off your gas furnace automatically. Also with the help of motion detectors it improves Nest Thermostat’s auto-away functionality. It looks elegant and costs a lot more than a regular smoke detector Just like Nest Thermostat, Nest Protect is elegant and adorable. You just fall in love with it the moment you see it. It’s another master piece from the designer of Apple’s iPod. All is good with the Nest Protect, except the price!! It costs whooping $129, which is almost 4 times more expensive than the best selling conventional thermostats available at $30. A single bed room apartment would require at least 3 detectors and it costs around $390 to install Nest Protects compared to 90$ required for conventional smoke detectors. Though Nest Thermostat is an expensive one compared to conventional thermostats, it offered great savings through its intelligent auto-away feature. Users were able to able to see returns on their investments. If Nest Protect also can provide good return on investment the it will be very successful.

    Read the article

  • Sunshine after the iCloud release?

    - by Laila
    "Why should I believe them? They're the ones that brought us MobileMe? It was not our finest hour, but we learned a lot." Steve Jobs June 6th 2011 Apple's new cloud service has been met with uncritical excitement by industry commentators.  It is wonderful what a rename can do.  Apple has had a 'cloud' offering for three years called MobileMe, successor to .MAC and  iTools, so iCloud is now the fourth internet service Apple have attempted. If this had been Microsoft, there would have been catcalls all around the blogosphere.  I'll admit that there is a lot more functionality announced for iCloud than MobileMe has ever managed to achieve, but then almost anything has more functionality than MobileMe.  It's an expensive service (£120 a year in the UK, $90 in the states), launched as far back as  June 9, 2008, that has delivered very little and suffered a string of technical problems; the documentation was mainly  a community effort, built up gradually by the frustrated and angry users. It was supposed to synchronise PC Outlook calendars but couldn't manage Microsoft Exchange (Google could, of course). It used WebDAV to allow Windows users to attach to the filestore, but didn't document how to do it. The method for downloading and uploading files to the cloud-based filestore was ridiculously clunky. It allowed you to post photos on a public site, but forgot to include a way of deleting photos. I could go on with the list, but you can explore the many sites that have flourished to inhabit the support-vacuum left by Apple. MobileMe should have had all the bright new clever things announced for iCloud. Apple dropped the ball, and allowed services such as Flickr to fill the void. However, their PR skills are such that, a name-change later (the .ME.com email address remains), it has turned a rout into a victory, and hundreds of earnest bloggers have been extolling Apple's expertise in cloud matters. This must be frustrating for the other cloud providers who have quietly got the technology working right. I wish iCloud well, even though I resent the expensive mess they made of MobileMe. Apple promise that iCloud will sync files, apps, app data, and media across all the different iOS5 devices, Macs, and PCs. It also hopes to sync music across devices, but not video content. They've offered existing MobileMe users free use of the MobileMe service for a year as the product is morphed, and they will be able to transfer to iCloud when it is launched in the autumn.  On June 30, 2012, MobileMe will die, and Apple's iWeb is also soon to join iTools and .MAC in the hereafter. So why get excited about iCloud? That all depends on the level of PC integration. Whereas iOS5 machines will be full participants in the new world of data-sharing (Sorry iPod Touch users) what about .NET libraries? There is talk of synchronising 'My Pictures' libraries with iOS5 and iMac machines, but little more detail as yet. Apple has a lot to prove with iCloud and anyone with actual experience of their past attempts to get into cloud services will be wary.

    Read the article

  • Efficiently separating Read/Compute/Write steps for concurrent processing of entities in Entity/Component systems

    - by TravisG
    Setup I have an entity-component architecture where Entities can have a set of attributes (which are pure data with no behavior) and there exist systems that run the entity logic which act on that data. Essentially, in somewhat pseudo-code: Entity { id; map<id_type, Attribute> attributes; } System { update(); vector<Entity> entities; } A system that just moves along all entities at a constant rate might be MovementSystem extends System { update() { for each entity in entities position = entity.attributes["position"]; position += vec3(1,1,1); } } Essentially, I'm trying to parallelise update() as efficiently as possible. This can be done by running entire systems in parallel, or by giving each update() of one system a couple of components so different threads can execute the update of the same system, but for a different subset of entities registered with that system. Problem In reality, these systems sometimes require that entities interact(/read/write data from/to) each other, sometimes within the same system (e.g. an AI system that reads state from other entities surrounding the current processed entity), but sometimes between different systems that depend on each other (i.e. a movement system that requires data from a system that processes user input). Now, when trying to parallelize the update phases of entity/component systems, the phases in which data (components/attributes) from Entities are read and used to compute something, and the phase where the modified data is written back to entities need to be separated in order to avoid data races. Otherwise the only way (not taking into account just "critical section"ing everything) to avoid them is to serialize parts of the update process that depend on other parts. This seems ugly. To me it would seem more elegant to be able to (ideally) have all processing running in parallel, where a system may read data from all entities as it wishes, but doesn't write modifications to that data back until some later point. The fact that this is even possible is based on the assumption that modification write-backs are usually very small in complexity, and don't require much performance, whereas computations are very expensive (relatively). So the overhead added by a delayed-write phase might be evened out by more efficient updating of entities (by having threads work more % of the time instead of waiting). A concrete example of this might be a system that updates physics. The system needs to both read and write a lot of data to and from entities. Optimally, there would be a system in place where all available threads update a subset of all entities registered with the physics system. In the case of the physics system this isn't trivially possible because of race conditions. So without a workaround, we would have to find other systems to run in parallel (which don't modify the same data as the physics system), other wise the remaining threads are waiting and wasting time. However, that has disadvantages Practically, the L3 cache is pretty much always better utilized when updating a large system with multiple threads, as opposed to multiple systems at once, which all act on different sets of data. Finding and assembling other systems to run in parallel can be extremely time consuming to design well enough to optimize performance. Sometimes, it might even not be possible at all because a system just depends on data that is touched by all other systems. Solution? In my thinking, a possible solution would be a system where reading/updating and writing of data is separated, so that in one expensive phase, systems only read data and compute what they need to compute, and then in a separate, performance-wise cheap, write phase, attributes of entities that needed to be modified are finally written back to the entities. The Question How might such a system be implemented to achieve optimal performance, as well as making programmer life easier? What are the implementation details of such a system and what might have to be changed in the existing EC-architecture to accommodate this solution?

    Read the article

  • Investigating on xVelocity (VertiPaq) column size

    - by Marco Russo (SQLBI)
      In January I published an article about how to optimize high cardinality columns in VertiPaq. In the meantime, VertiPaq has been rebranded to xVelocity: the official name is now “xVelocity in-memory analytics engine (VertiPaq)” but using xVelocity and VertiPaq when we talk about Analysis Services has the same meaning. In this post I’ll show how to investigate on columns size of an existing Tabular database so that you can find the most important columns to be optimized. A first approach can be looking in the DataDir of Analysis Services and look for the folder containing the database. Then, look for the biggest files in all subfolders and you will find the name of a file that contains the name of the most expensive column. However, this heuristic process is not very optimized. A better approach is using a DMV that provides the exact information. For example, by using the following query (open SSMS, open an MDX query on the database you are interested to and execute it) you will see all database objects sorted by used size in a descending way. SELECT * FROM $SYSTEM.DISCOVER_STORAGE_TABLE_COLUMN_SEGMENTS ORDER BY used_size DESC You can look at the first rows in order to understand what are the most expensive columns in your tabular model. The interesting data provided are: TABLE_ID: it is the name of the object – it can be also a dictionary or an index COLUMN_ID: it is the column name the object belongs to – you can also see ID_TO_POS and POS_TO_ID in case they refer to internal indexes RECORDS_COUNT: it is the number of rows in the column USED_SIZE: it is the used memory for the object By looking at the ration between USED_SIZE and RECORDS_COUNT you can understand what you can do in order to optimize your tabular model. Your options are: Remove the column. Yes, if it contains data you will never use in a query, simply remove the column from the tabular model Change granularity. If you are tracking time and you included milliseconds but seconds would be enough, round the data source column to the nearest second. If you have a floating point number but two decimals are good enough (i.e. the temperature), round the number to the nearest decimal is relevant to you. Split the column. Create two or more columns that have to be combined together in order to produce the original value. This technique is described in VertiPaq optimization article. Sort the table by that column. When you read the data source, you might consider sorting data by this column, so that the compression will be more efficient. However, this technique works better on columns that don’t have too many distinct values and you will probably move the problem to another column. Sorting data starting from the lower density columns (those with a few number of distinct values) and going to higher density columns (those with high cardinality) is the technique that provides the best compression ratio. After the optimization you should be able to reduce the used size and improve the count/size ration you measured before. If you are interested in a longer discussion about internal storage in VertiPaq and you want understand why this approach can save you space (and time), you can attend my 24 Hours of PASS session “VertiPaq Under the Hood” on March 21 at 08:00 GMT.

    Read the article

  • Comparing Apples and Pairs

    - by Tony Davis
    A recent study, High Costs and Negative Value of Pair Programming, by Capers Jones, pulls no punches in its assessment of the costs-to- benefits ratio of pair programming, two programmers working together, at a single computer, rather than separately. He implies that pair programming is a method rushed into production on a wave of enthusiasm for Agile or Extreme Programming, without any real regard for its effectiveness. Despite admitting that his data represented a far from complete study of the economics of pair programming, his conclusions were stark: it was 2.5 times more expensive, resulted in a 15% drop in productivity, and offered no significant quality benefits. The author provides a more scientific analysis than Jon Evans’ Pair Programming Considered Harmful, but the theme is the same. In terms of upfront-coding costs, pair programming is surely more expensive. The claim of productivity loss is dubious and contested by other studies. The third claim, though, did surprise me. The author’s data suggests that if both the pair and the individual programmers employ static code analysis and testing, then there is no measurable difference in the resulting code quality, in terms of defects per function point. In other words, pair programming incurs a massive extra cost for no tangible return in investment. There were, inevitably, many criticisms of his data and his conclusions, a few of which are persuasive. Firstly, that the driver/observer model of pair programming, on which the study bases its findings, is far from the most effective. For example, many find Ping-Pong pairing, based on use of test-driven development, far more productive. Secondly, that it doesn’t distinguish between “expert” and “novice” pair programmers– that is, independently of other programming skills, how skilled was an individual at pair programming. Thirdly, that his measure of quality is too narrow. This point rings true, certainly at Red Gate, where developers don’t pair program all the time, but use the method in short bursts, while tackling a tricky problem and needing a fresh perspective on the best approach, or more in-depth knowledge in a particular domain. All of them argue that pair programming, and collective code ownership, offers significant rewards, if not in terms of immediate “bug reduction”, then in removing the likelihood of single points of failure, and improving the overall quality and longer-term adaptability/maintainability of the design. There is also a massive learning benefit for both participants. One developer told me how he once worked in the same team over consecutive summers, the first time with no pair programming and the second time pair-programming two-thirds of the time, and described the increased rate of learning the second time as “phenomenal”. There are a great many theories on how we should develop software (Scrum, XP, Lean, etc.), but woefully little scientific research in their effectiveness. For a group that spends so much time crunching other people’s data, I wonder if developers spend enough time crunching data about themselves. Capers Jones’ data may be incomplete, but should cause a pause for thought, especially for any large IT departments, supporting commerce and industry, who are considering pair programming. It certainly shouldn’t discourage teams from exploring new ways of developing software, as long as they also think about how to gather hard data to gauge their effectiveness.

    Read the article

  • Investigating on xVelocity (VertiPaq) column size

    - by Marco Russo (SQLBI)
      In January I published an article about how to optimize high cardinality columns in VertiPaq. In the meantime, VertiPaq has been rebranded to xVelocity: the official name is now “xVelocity in-memory analytics engine (VertiPaq)” but using xVelocity and VertiPaq when we talk about Analysis Services has the same meaning. In this post I’ll show how to investigate on columns size of an existing Tabular database so that you can find the most important columns to be optimized. A first approach can be looking in the DataDir of Analysis Services and look for the folder containing the database. Then, look for the biggest files in all subfolders and you will find the name of a file that contains the name of the most expensive column. However, this heuristic process is not very optimized. A better approach is using a DMV that provides the exact information. For example, by using the following query (open SSMS, open an MDX query on the database you are interested to and execute it) you will see all database objects sorted by used size in a descending way. SELECT * FROM $SYSTEM.DISCOVER_STORAGE_TABLE_COLUMN_SEGMENTS ORDER BY used_size DESC You can look at the first rows in order to understand what are the most expensive columns in your tabular model. The interesting data provided are: TABLE_ID: it is the name of the object – it can be also a dictionary or an index COLUMN_ID: it is the column name the object belongs to – you can also see ID_TO_POS and POS_TO_ID in case they refer to internal indexes RECORDS_COUNT: it is the number of rows in the column USED_SIZE: it is the used memory for the object By looking at the ration between USED_SIZE and RECORDS_COUNT you can understand what you can do in order to optimize your tabular model. Your options are: Remove the column. Yes, if it contains data you will never use in a query, simply remove the column from the tabular model Change granularity. If you are tracking time and you included milliseconds but seconds would be enough, round the data source column to the nearest second. If you have a floating point number but two decimals are good enough (i.e. the temperature), round the number to the nearest decimal is relevant to you. Split the column. Create two or more columns that have to be combined together in order to produce the original value. This technique is described in VertiPaq optimization article. Sort the table by that column. When you read the data source, you might consider sorting data by this column, so that the compression will be more efficient. However, this technique works better on columns that don’t have too many distinct values and you will probably move the problem to another column. Sorting data starting from the lower density columns (those with a few number of distinct values) and going to higher density columns (those with high cardinality) is the technique that provides the best compression ratio. After the optimization you should be able to reduce the used size and improve the count/size ration you measured before. If you are interested in a longer discussion about internal storage in VertiPaq and you want understand why this approach can save you space (and time), you can attend my 24 Hours of PASS session “VertiPaq Under the Hood” on March 21 at 08:00 GMT.

    Read the article

  • Can someone explain the true landscape of Rails vs PHP deployment, particularly within the context of Reseller-based web hosting (e.g., Hostgator)?

    - by rcd
    Currently, I have a reseller account with the company HostGator. I design websites, which up until now have occasionally been wrapped in Wordpress CMSs and the like (PHP applications). I then sell hosting (of the site I've designed) to the client, which is pretty simple, in that I can simply click a button and add a new shared hosting account/site with whatever settings I want. Furthermore, I then utilize WHMCS to automate billing and account management. It's a nice package and pretty simple. I pay something like $25 a month, and can sell a hundred accounts under this (because my clients bandwidth requirements are low). Now I am finding the need to develop more customized applications, including a minimalist CMS and several proprietary things. I soon anticipate developing these apps for clients as well. Thus, I've spent the past few months learning Rails, and it's coming along well now. The thing that has nagged at me all along, though, is the deployment issue. I can't wrap my brain around it. It seems like all of the popular options (Heroku, etc) have nice automation with git and are set up in the "Rails Way". I get that (sort of). But it's terribly expensive... a single dyno, a helper, and the cheapest database (which they say is mainly suitable for testing) that isn't limited to 5MB runs $51. This is for ONE app!!! Throw in a "production" DB and you're over $200. This is like... the same prices as getting a server somewhere, right? Meanwhile, going back to what I guess is a "traditional" hosting environment with Hostgator, their server only has Ruby 1.8.7 and Rails 2.3.5... No Rails 3. AND, no Passenger (not that I really understand the difference in CGI or mod_rails or whatever, but they say Passenger is the simplest). So I'm to understand that if I build an app in Rails 3, it won't run at all on this host? But damn, I already have these accounts under my reseller account there, all running static html and/or PHP stuff, right? So what now? How do I get all of this under one simple (and affordable) roof? Forgive my ignorance, but I just don't get it. Managing a VPS is cool and all, but entails learning server admin stuff and security... And it's expensive. I get that a shared and/or reseller "server-based" (forgive the terminology) may be inadequate for large-scale apps that use a lot of bandwidth... But what about for those of us who are building real (but small and low bandwidth) apps (with Rails) and who want to deploy them simply, cheaply, using the same conceptual approach as PHP? Even after learning all of this Ruby and Rails stuff for months, I'm questioning whether it's worth it when it comes to deployment. I want to build a small app, upload it to my home directory on a shared server account, and just make it run. Why should that be so hard? Am I just choosing the wrong language/framework? Forgive my ignorance in the subject; these questions are not rhetorical; just trying to learn here. So: 1) I'd appreciate if someone could give me a good rundown of how to understand deployment in Rails vs. PHP. 2) I'd appreciate if someone could address my issue with running a hosting/web business around reseller hosting (Hostgator) while also being able to host Rails apps. Can it be done? And how can a company like Hostgator completely ignore what's current in Rails/Ruby? Thanks.

    Read the article

  • latex list environment inside the tabular environment: extra line at top preventing alignment

    - by Usagi
    Hello good people of stackoverflow. I have a LaTeX question that is bugging me. I have been trying to get a list environment to appear correctly inside the tabular environment. So far I have gotten everything to my liking except one thing: the top of the list does not align with other entries in the table, in fact it looks like it adds one line above the list... I would like to have these lists at the top. This is what I have, a custom list environment: \newenvironment{flushemize}{ \begin{list}{$\bullet$} {\setlength{\itemsep}{1pt} \setlength{\parskip}{0pt} \setlength{\parsep}{0pt} \setlength{\partopsep}{0pt} \setlength{\topsep}{0pt} \setlength{\leftmargin}{12pt}}}{\end{list}} Renamed ragged right: \newcommand{\rr}{\raggedright} and here is my table: \begin{table}[H]\caption{Tank comparisons}\label{tab:tanks} \centering \rowcolors{2}{white}{tableShade} \begin{tabular}{p{1in}p{1.5in}p{1.5in}rr} \toprule {\bf Material} & {\bf Pros} & {\bf Cons} & {\bf Size} & {\bf Cost} \\ \midrule \rr Reinforced concrete &\rr \begin{flushemize}\item Strong \item Secure \end{flushemize}&\rr \begin{flushemize}\item Prone to leaks \item Relatively expensive to install \item Heavy \end{flushemize} & 100,000 gal & \$299,400 \\ \rr Steel & \begin{flushemize}\item Strong \item Secure \end{flushemize} & \begin{flushemize}\item Relatively expensive to install \item Heavy \item Require painting to prevent rusting \end{flushemize} & 100,000 gal & \$130,100 \\ \rr Polypropylene & \begin{flushemize}\item Easy to install \item Mobile \item Inexpensive \item Prefabricated \end{flushemize} & \begin{flushemize}\item Relatively insecure \item Max size available 10,000 gal \end{flushemize} & 10,000 gal & \$5,000 \\ \rr Wood & \begin{flushemize}\item Easy to install \item Mobile \item Cheap to install \end{flushemize} & \begin{flushemize}\item Prone to rot \item Must remain full once constructed \end{flushemize} & 100,000 gal & \$86,300\\ \bottomrule \end{tabular} \end{table} Thank you for any advice :)

    Read the article

  • ai: Determining what tests to run to get most useful data

    - by Sai Emrys
    This is for http://cssfingerprint.com I have a system (see about page on site for details) where: I need to output a ranked list, with confidences, of categories that match a particular feature vector the binary feature vectors are a list of site IDs & whether this session detected a hit feature vectors are, for a given categorization, somewhat noisy (sites will decay out of history, and people will visit sites they don't normally visit) categories are a large, non-closed set (user IDs) my total feature space is approximately 50 million items (URLs) for any given test, I can only query approx. 0.2% of that space I can only make the decision of what to query, based on results so far, ~10-30 times, and must do so in <~100ms (though I can take much longer to do post-processing, relevant aggregation, etc) getting the AI's probability ranking of categories based on results so far is mildly expensive; ideally the decision will depend mostly on a few cheap sql queries I have training data that can say authoritatively that any two feature vectors are the same category but not that they are different (people sometimes forget their codes and use new ones, thereby making a new user id) I need an algorithm to determine what features (sites) are most likely to have a high ROI to query (i.e. to better discriminate between plausible-so-far categories [users], and to increase certainty that it's any given one). This needs to take into balance exploitation (test based on prior test data) and exploration (test stuff that's not been tested enough to find out how it performs). There's another question that deals with a priori ranking; this one is specifically about a posteriori ranking based on results gathered so far. Right now, I have little enough data that I can just always test everything that anyone else has ever gotten a hit for, but eventually that won't be the case, at which point this problem will need to be solved. I imagine that this is a fairly standard problem in AI - having a cheap heuristic for what expensive queries to make - but it wasn't covered in my AI class, so I don't actually know whether there's a standard answer. So, relevant reading that's not too math-heavy would be helpful, as well as suggestions for particular algorithms. What's a good way to approach this problem?

    Read the article

  • Conditions with common logic: question of style, readability, efficiency, ...

    - by cdonner
    I have conditional logic that requires pre-processing that is common to each of the conditions (instantiating objects, database lookups etc). I can think of 3 possible ways to do this, but each has a flaw: Option 1 if A prepare processing do A logic else if B prepare processing do B logic else if C prepare processing do C logic // else do nothing end The flaw with option 1 is that the expensive code is redundant. Option 2 prepare processing // not necessary unless A, B, or C if A do A logic else if B do B logic else if C do C logic // else do nothing end The flaw with option 2 is that the expensive code runs even when neither A, B or C is true Option 3 if (A, B, or C) prepare processing end if A do A logic else if B do B logic else if C do C logic end The flaw with option 3 is that the conditions for A, B, C are being evaluated twice. The evaluation is also costly. Now that I think about it, there is a variant of option 3 that I call option 4: Option 4 if (A, B, or C) prepare processing if A set D else if B set E else if C set F end end if D do A logic else if E do B logic else if F do C logic end While this does address the costly evaluations of A, B, and C, it makes the whole thing more ugly and I don't like it. How would you rank the options, and are there any others that I am not seeing?

    Read the article

  • Python SQLite FTS3 alternatives?

    - by Mike Cialowicz
    Are there any good alternatives to SQLite + FTS3 for python? I'm iterating over a series of text documents, and would like to categorize them according to some text queries. For example, I might want to know if a document mentions the words "rating" or "upgraded" within three words of "buy." The FTS3 syntax for this query is the following: (rating OR upgraded) NEAR/3 buy That's all well and good, but if I use FTS3, this operation seems rather expensive. The process goes something like this: # create an SQLite3 db in memory conn = sqlite3.connect(':memory:') c = conn.cursor() c.execute('CREATE VIRTUAL TABLE fts USING FTS3(content TEXT)') conn.commit() Then, for each document, do something like this: #insert the document text into the fts table, so I can run a query c.execute('insert into fts(content) values (?)', content) conn.commit() # execute my FTS query here, look at the results, etc # remove the document text from the fts table before working on the next document c.execute('delete from fts') conn.commit() This seems rather expensive to me. The other problem I have with SQLite FTS is that it doesn't appear to work with Python 2.5.4. The 'CREATE VIRTUAL TABLE' syntax is unrecognized. This means that I'd have to upgrade to Python 2.6, which means re-testing numerous existing scripts and programs to make sure they work under 2.6. Is there a better way? Perhaps a different library? Something faster? Thank you.

    Read the article

  • Legitimate uses of the Function constructor

    - by Marcel Korpel
    As repeatedly said, it is considered bad practice to use the Function constructor (also see the ECMAScript Language Specification, 5th edition, § 15.3.2.1): new Function ([arg1[, arg2[, … argN]],] functionBody) (where all arguments are strings containing argument names and the last (or only) string contains the function body). To recapitulate, it is said to be slow, as explained by the Opera team: Each time […] the Function constructor is called on a string representing source code, the script engine must start the machinery that converts the source code to executable code. This is usually expensive for performance – easily a hundred times more expensive than a simple function call, for example. (Mark ‘Tarquin’ Wilton-Jones) Though it's not that bad, according to this post on MDC (I didn't test this myself using the current version of Firefox, though). Crockford adds that [t]he quoting conventions of the language make it very difficult to correctly express a function body as a string. In the string form, early error checking cannot be done. […] And it is wasteful of memory because each function requires its own independent implementation. Another difference is that a function defined by a Function constructor does not inherit any scope other than the global scope (which all functions inherit). (MDC) Apart from this, you have to be attentive to avoid injection of malicious code, when you create a new Function using dynamic contents. Lots of disadvantages and it is intelligible that ECMAScript 5 discourages the use of the Function constructor by throwing an exception when using it in strict mode (§ 13.1). That said, T.J. Crowder says in an answer that [t]here's almost never any need for the similar […] new Function(...), either, again except for some advanced edge cases. So, now I am wondering: what are these “advanced edge cases”? Are there legitimate uses of the Function constructor?

    Read the article

  • Help with MySQL query... Need help ordering a group of rows

    - by user156814
    I can tell it best by explaining the query I have, and what I need. I need to be able to get a group of items from the database, grouped by category, manufacturer, and year made. The groupings need to be sorted based on total amount of items within the group. This part is done with the query below. Secondly, I need to be able to show an image of the most expensive item out of the group, which is why I use MAX(items.current_price). I thought MAX() gets the ENTIRE row corresponding to the largest column value. I was wrong, as MAX only gets the numeric value of the largest price. So the query doesnt work well for that. SELECT items.id, items.year, items.manufacturer, COUNT(items.id) AS total, MAX(items.current_price) AS price, items.gallery_url, FROM ebay AS items WHERE items.primary_category_id = 213 AND items.year <> '' AND items.manufacturer <> '' AND items.bad_item <> 1 GROUP BY items.primary_category_id, items.manufacturer, items.year ORDER BY total DESC, price ASC LIMIT 10 if that doesnt explain it well, the results should be something like this id 10548 year 1989 manufacturer bowman total 451 price 8500.00 (The price of the most expensive item in the table/ not the price of item 10548) gallery_url http://ebay.xxxxx (The image of item 10548) A little help please. Thanks

    Read the article

  • Unmanaged Process in Mono

    - by Residuum
    I want to start a quite expensive process (jackd) from a Mono application, and do not need full access to the process from the application itself. As the process is so expensive in terms of CPU usage, a Glib.IdleHandler for polling the process will not work, as it is never executed, and the GUI becomes unresponsive. Is there any way to have the cake and eating it at the same time in Mono? EDIT: I only need to be able to start and stop the process from Mono, I do not need information about the state of the process or if it has exited, as my application will register itself as a client to jackd, basically I need a "replacement" for bash's jackd &>/dev/null 2>&1 & for the System.Diagnostics.Process ;). Here is what I have so far for starting and stopping the process: public void StartJackd() { _jackd = new Process (); _jackd.StartInfo = _jackdStartup; if (_jackd.Start ()) { _jackd.EnableRaisingEvents = true; _jackd.Exited += JackdExited; } } public void StopJackd() { if (_jackd != null && !_jackd.HasExited) { _jackd.CloseMainWindow (); } } And somewhere else I have this code for registering the IdleHandler: GLib.Idle.Add(new GLib.IdleHandler(UpdateJackdConnections)); This handler will fire all the time, while the process is not running, but never, when jackd is running.

    Read the article

  • Is locking on the requested object a bad idea?

    - by Quick Joe Smith
    Most advice on thread safety involves some variation of the following pattern: public class Thing { private static readonly object padlock = new object(); private string stuff, andNonsense; public string Stuff { get { lock (Thing.padlock) { if (this.stuff == null) this.stuff = "Threadsafe!"; } return this.stuff; } } public string AndNonsense { get { lock (Thing.padlock) { if (this.andNonsense == null) this.andNonsense = "Also threadsafe!"; } return this.andNonsense; } } // Rest of class... } In cases where the get operations are expensive and unrelated, a single locking object is unsuitable because a call to Stuff would block all calls to AndNonsense, degrading performance. And rather than create a lock object for each call, wouldn't it be better to acquire the lock on the member itself (assuming it is not something that implements SyncRoot or somesuch for that purpose? For example: public string Stuff { get { lock (this.stuff) { // Pretend that this is a very expensive operation. if (this.stuff == null) this.stuff = "Still threadsafe and good?"; } return this.stuff; } } Strangely, I have never seen this approach recommended or warned against. Am I missing something obvious?

    Read the article

  • Are background threads a bad idea? Why?

    - by Matt Grande
    So I've been told what I'm doing here is wrong, but I'm not sure why. I have a webpage that imports a CSV file with document numbers to perform an expensive operation on. I've put the expensive operation into a background thread to prevent it from blocking the application. Here's what I have in a nutshell. protected void ButtonUpload_Click(object sender, EventArgs e) { if (FileUploadCSV.HasFile) { string fileText; using (var sr = new StreamReader(FileUploadCSV.FileContent)) { fileText = sr.ReadToEnd(); } var documentNumbers = fileText.Split(new[] {',', '\n', '\r'}, StringSplitOptions.RemoveEmptyEntries); ThreadStart threadStart = () => AnotherClass.ExpensiveOperation(documentNumbers); var thread = new Thread(threadStart) {IsBackground = true}; thread.Start(); } } (obviously with some error checking & messages for users thrown in) So my three-fold question is: a) Is this a bad idea? b) Why is this a bad idea? c) What would you do instead?

    Read the article

  • Caching vector addition over changing collections

    - by DRMacIver
    I have the following setup: I have a largish number of uuids (currently about 10k but expected to grow unboundedly - they're user IDs) and a function f : id - sparse vector with 32-bit integer values (no need to worry about precision). The function is reasonably expensive (not outrageously so, but probably on the order of a few 100ms for a given id). The dimension of the sparse vectors should be assumed to be infinite, as new dimensions can appear over time, but in practice is unlikely to ever exceed about 20k (and individual results of f are unlikely to have more than a few hundred non-zero values). I want to support the following operations efficiently: add a new ID to the collection invalidate an existing ID retrieve sum f(id) in O(changes since last retrieval) i.e. I want to cache the sum of the vectors in a way that's reasonable to do incrementally. One option would be to support a remove ID operation and treat invalidation as a remove followed by an add. The problem with this is that it requires us to keep track of all the old values of f, which is expensive in space. I potentially need to use many instances of this sort of cached structure, so I would like to avoid that. The likely usage pattern is that new IDs are added at a fairly continuous rate and are frequently invalidated at first. Ids which have been invalidated recently are much more likely to be invalidated again than ones which have remained valid for a long time, but in principle an old Id can still be invalidated. Ideally I don't want to do this in memory (or at least I want a way that lets me save the result to disk efficiently), so an idea which lets me piggyback off an existing DB implementation of some sort would be especially appreciated.

    Read the article

  • Understanding the value of Customer Experience & Loyalty for the Telecommunications Industry

    - by raul.goycoolea
    Worried by economic woes and market forces, especially in mature markets, communications service providers (CSPs) increasingly focus on improving customer experience. In fact, it seems difficult to find a major message by a C-level executive in the developed world that does not include something on "meeting and exceeding customers' needs". Frequently in customer satisfaction studies by prominent firms, CSPs fall short of the leadership demonstrated by other industries that take customer-centric approaches to their bottom-line strategies. Consider the following:Despite the continued impact of global economic crisis, in July 2010, Apple Computer posted record revenue and net quarterly profit. Those who attribute the results primarily to the iPhone 4 launch should note that Apple also shipped around 30% more Macintosh computers than the same period the previous year. Even sales of the iPod line increased by 8% in a highly commoditized, shrinking media player market. Finally, Apple began selling iPads during the quarter, with total sales of more than 3 million units. What does Apple have that the others lack? Well, some great products (and services) to be sure, but it also excels at customer service and support, marketing, and distribution, and has one of the strongest brands globally. Its products are useful, simple to use, easy to acquire and augment, high quality, and considered very cool. They also evoke such an emotional response from many of Apple's customers, which they turn up their noses at competitive products.In other words, Apple appears to have mastered virtually every aspect of customer experience and the resultant loyalty of its customer base - even in difficult financial times. Through that unwavering customer focus, Apple continues to drive its revenues and profits to new heights. Other customer loyalty leaders like Wal-Mart, Google, Toyota and Honda are also doing well by focusing on customer experience as an essential driver of profitability. Service providers should note this performance and ask themselves how they might leverage the same principles to increase their own profitability. After all, that is what customer experience and loyalty are all about: profitability.To successfully manage all the critical touch points of customer experience, CSPs must shun the one-size-fits-all approach. They can no longer afford to view customer service fundamentally as an act of altruism - which mentality dates back to the industry's civil service days, when CSPs were typically government organizations that were critical to economic development and public safety.As regulators and public officials have pushed, and continue to push, service providers to new heights of reliability - using incentives and punishments - most CSPs already have some of the fundamental building blocks of customer service in place. Yet despite that history and experience, service providers still lag other industries in providing what is seen as good customer service.As we observed in the TMF's 2009 Insights Research report, Customer Experience Management: Driving Loyalty & Profitability there has been resurgence in interest by CSPs. More and more of them have stated ambitions to catch up other industries, and they are realizing that good customer service is a powerful strategy for increasing business performance and profitability, not an act of good will.CSPs are recognizing the connection between customer experience and profitability, as demonstrated in many studies. For example, according to research by Bain & Company, a 5 percent improvement in customer retention rates can yield as much as a 75 percent increase in profits for companies across a range of industries.After decades of customer experience strategy formulation, Bain partner and business author, Frederick Reichheld, considers "would you recommend us to a friend?" as the ultimate question for a customer. How many times have you or your friends recommended an iPod, iPhone or a Mac? What do your children recommend to their peers? Their peers to them?There are certain steps service providers have to take to create more personalized relationships with their customers, as well as reduce churn and increase profitability, all while becoming leaner and more agile. First, they have to define customer experience, we define it as the result of the sum of observations, perceptions, thoughts and feelings arising from interactions and relationships between customers and their service provider(s). Virtually every customer touch point - whether directly or indirectly linked to service providers and their partners - contributes to customer perception, satisfaction, loyalty, and ultimately profitability. Gaining leadership in customer experience and satisfaction will not be a simple task, as it is affected by virtually every customer-facing aspect of the service provider, and in turn impacts the service provider deeply - especially on the all-important bottom line. The scope of issues affecting customer experience is complex and dynamic.With new services, devices and applications extending the basis of customer experience to domains beyond the direct control of the service provider, it is likely to increase in complexity and dynamism.Customer loyalty = increased profitsAs stated earlier, customer experience programs are not fundamentally altruistic exercises, but a strategic means of improving competitiveness and profitability in the short and long term. Loyalty is essential to deriving long term profits from customers.Some of the earliest loyalty programs date back to the 1930s, when packaged goods companies offered embedded coupons for rewards to buyers, and eventually retail chains began offering reward programs to frequent shoppers. These programs continued for decades but were leapfrogged in the 1980s by more aggressive programs from the airlines.This movement was led by American Airlines, which launched the first full-scale loyalty marketing program of the modern era with the AAdvantage frequent flyer scheme. It was the first to reward frequent fliers with notional air miles that could be accumulated and later redeemed for free travel. Figure 1: Opportunities example of Customer loyalty driven profitOther airlines and travel providers were quick to grasp the incredible value of providing customers with an incentive to use their company exclusively. Within a few years, dozens of travel industry companies launched similar initiatives and now loyalty programs are achieving near-ubiquity in many service industries, especially those in which it is difficult to differentiate offerings by product attributes.The belief is that increased profitability will result from customer retention efforts because:•    The cost of acquisition occurs only at the beginning of a relationship: the longer the relationship, the lower the amortized cost;•    Account maintenance costs decline as a percentage of total costs, or as a percentage of revenue, over the lifetime of the relationship;•    Long term customers tend to be less inclined to switch and less price sensitive which can result in stable unit sales volume and increases in dollar-sales volume;•    Long term customers may initiate word-of-mouth promotions and referrals, which cost the company nothing and arguably are the most effective form of advertising;•    Long-term customers are more likely to buy ancillary products and higher margin supplemental products;•    Long term customers tend to be satisfied with their relationship with the company and are less likely to switch to competitors, making market entry or competitors gaining market share difficult;•    Regular customers tend to be less expensive to service, as they are familiar with the processes involved, require less 'education', and are consistent in their order placement;•    Increased customer retention and loyalty makes the employees' jobs easier and more satisfying. In turn, happy employees feed back into higher customer satisfaction in a virtuous circle. Figure 2: The virtuous circle of customer loyaltyFigure 2 represents a high-level example of a virtuous cycle driven by customer satisfaction and loyalty, depicting how superiority in product and service offerings, as well as strong customer support by competent employees, lead to higher sales and ultimately profitability. As stated above, this is not a new concept, but succeeding with it is difficult. It has eluded many a company driven to achieve profitability goals. Of course, for this circle to be virtuous, the customer relationship(s) must be profitable.Trying to maintain the loyalty of unprofitable customers is not a viable business strategy. It is, therefore, important that marketers can assess the profitability of each customer (or customer segment), and either improve or terminate relationships that are not profitable. This means each customer's 'relationship costs' must be understood and compared to their 'relationship revenue'. Customer lifetime value (CLV) is the most commonly used metric here, as it is generally accepted as a representation of exactly how much each customer is worth in monetary terms, and therefore a determinant of exactly how much a service provider should be willing to spend to acquire or retain that customer.CLV models make several simplifying assumptions and often involve the following inputs:•    Churn rate represents the percentage of customers who end their relationship with a company in a given period;•    Retention rate is calculated by subtracting the churn rate percentage from 100;•    Period/horizon equates to the units of time into which a customer relationship can be divided for analysis. A year is the most commonly used period for this purpose. Customer lifetime value is a multi-period calculation, often projecting three to seven years into the future. In practice, analysis beyond this point is viewed as too speculative to be reliable. The model horizon is the number of periods used in the calculation;•    Periodic revenue is the amount of revenue collected from a customer in a given period (though this is often extended across multiple periods into the future to understand lifetime value), such as usage revenue, revenues anticipated from cross and upselling, and often some weighting for referrals by a loyal customer to others; •    Retention cost describes the amount of money the service provider must spend, in a given period, to retain an existing customer. Again, this is often forecast across multiple periods. Retention costs include customer support, billing, promotional incentives and so on;•    Discount rate means the cost of capital used to discount future revenue from a customer. Discounting is an advanced method used in more sophisticated CLV calculations;•    Profit margin is the projected profit as a percentage of revenue for the period. This may be reflected as a percentage of gross or net profit. Again, this is generally projected across the model horizon to understand lifetime value.A strong focus on managing these inputs can help service providers realize stronger customer relationships and profits, but there are some obstacles to overcome in achieving accurate calculations of CLV, such as the complexity of allocating costs across the customer base. There are many costs that serve all customers which must be properly allocated across the base, and often a simple proportional allocation across the whole base or a segment may not accurately reflect the true cost of serving that customer;  This is made worse by the fragmentation of customer information, which is likely to be across a variety of product or operations groups, and may be difficult to aggregate due to different representations.In addition, there is the complexity of account relationships and structures to take into consideration. Complex account structures may not be understood or properly represented. For example, a profitable customer may have a separate account for a second home or another family member, which may appear to be unprofitable. If the service provider cannot relate the two accounts, CLV is not properly represented and any resultant cancellation of the apparently unprofitable account may result in the customer churning from the profitable one.In summary, if service providers are to realize strong customer relationships and their attendant profits, there must be a very strong focus on data management. This needs to be coupled with analytics that help business managers and those who work in customer-facing functions offer highly personalized solutions to customers, while maintaining profitability for the service provider. It's clear that acquiring new customers is expensive. Advertising costs, campaign management expenses, promotional service pricing and discounting, and equipment subsidies make a serious dent in a new customer's profitability. That is especially true given the rising subsidies for Smartphone users, which service providers hope will result in greater profits from profits from data services profitability in future.  The situation is made worse by falling prices and greater competition in mature markets.Customer acquisition through industry consolidation isn't cheap either. A North American service provider spent about $2,000 per subscriber in its acquisition of a smaller company earlier this year. While this has allowed it to leapfrog to become the largest mobile service provider in the country, it required a total investment of more than $28 billion (including assumption of the acquiree's debt).While many operating cost synergies clearly made this deal more attractive to the acquiring company, this is certainly an expensive way to acquire customers: the cost per subscriber in this case is not out of line with the prices others have paid for acquisitions.While growth by acquisition certainly increases overall revenues, it often creates tremendous challenges for profitability. Organic growth through increased customer loyalty and retention is a more effective driver of profit, as well as a stronger predictor of future profitability. Service providers, especially those in mature markets, are increasingly recognizing this and taking steps toward a creating a more personalized, flexible and satisfying experience for their customers.In summary, the clearest path to profitability for companies in virtually all industries is through customer retention and maximization of lifetime value. Service providers would do well to recognize this and focus attention on profitable customer relationships.

    Read the article

  • SCOM 2007 versus Zenoss (or other open source)

    - by TheCleaner
    I've taken the liberty to test both SCOM 2007 and Zenoss and found the following: SCOM 2007 Pros: Great MS Windows server monitoring and reporting In-depth configuration and easily integrates into a "MS datacenter" Cons: limited network device monitoring support (without 3rd party plugins) expensive difficult learning curve Zenoss Pros: Open Source (free) decent server monitoring for Windows, great monitoring for Linux decent network device monitoring Cons: not as in-depth as SCOM (for Windows at least) So my question to you folks is this: Given the above, and given that I'm trying to monitor 55 Windows servers, 1 Linux server, 2 ESX servers, and Juniper equipment...which would you recommend?

    Read the article

  • 6-core Sandy Bridge-E vs. 4-core Ivy Bridge

    - by Alexander Ilyin
    I am currently choosing between Intel Core i7-3770 (quad, Ivy) and Intel Core i7-3930K (6 cores, Sandy Bridge-E). This machine will be used for both work (Adobe, Autodesk software, graphic and coding-related) and gaming. Even if some applications I will use are capable to utilize all 6 cores at once, is it worth preferring Sandy Bridge-E to newer Ivy Bridge? Games aren't and probably will perform better on Ivy, won't they? 6-core is also twice as expensive as a quad Ivy.

    Read the article

  • Buying a Laptop Battery - OEM vs. 3rd Party

    - by pygorex1
    Looking at a replacement 9-cell battery for my Dell 1525 I've noticed that the OEM batteries that Dell sells are up to 3x more expensive than batteries sold by a 3rd party vendor. Is the Dell premium worth it? What experiences have you had buying replacement batteries?

    Read the article

< Previous Page | 7 8 9 10 11 12 13 14 15 16 17 18  | Next Page >