Search Results

Search found 15401 results on 617 pages for 'memory optimization'.

Page 33/617 | < Previous Page | 29 30 31 32 33 34 35 36 37 38 39 40  | Next Page >

• PHP-FPM High Memory Usage

  - by Ruel

  I have a wordpress blog, that uses WP-SuperCache, and normally I get 100 visitors per day. With nginx + php-fpm it's blazing fast, and I have no regrets. One thing i noticed, php-fpm takes a lot of memory: top - 09:20:43 up 5 days, 15:53, 1 user, load average: 0.00, 0.00, 0.00 Tasks: 26 total, 1 running, 25 sleeping, 0 stopped, 0 zombie Cpu(s): 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Mem: 1048576k total, 329956k used, 718620k free, 0k buffers Swap: 0k total, 0k used, 0k free, 0k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 10226 www-data 15 0 145m 52m 4584 S 0.0 5.1 0:07.55 php-fpm 10223 www-data 16 0 141m 48m 4692 S 0.0 4.8 0:08.70 php-fpm 20277 www-data 15 0 138m 46m 4368 S 0.0 4.5 0:07.55 php-fpm 20259 www-data 15 0 133m 41m 4600 S 0.0 4.0 0:06.68 php-fpm 12201 www-data 15 0 133m 41m 4632 S 0.0 4.0 0:08.31 php-fpm 11586 www-data 15 0 132m 40m 4292 S 0.0 3.9 0:03.27 php-fpm 29822 www-data 15 0 128m 36m 4356 S 0.0 3.6 0:05.26 php-fpm 28427 mysql 15 0 200m 7300 4764 S 0.0 0.7 0:47.89 mysqld 10202 root 18 0 98.3m 4320 1204 S 0.0 0.4 0:03.80 php-fpm 22524 root 18 0 86064 3396 2652 S 0.0 0.3 0:16.74 sshd 9882 www-data 18 0 42052 2572 804 S 0.0 0.2 0:27.52 nginx 9884 www-data 18 0 42052 2560 804 S 0.0 0.2 0:26.26 nginx 9881 www-data 18 0 42064 2524 804 S 0.0 0.2 0:29.24 nginx 9879 www-data 18 0 42032 2480 804 S 0.0 0.2 0:29.58 nginx 23771 root 15 0 12176 1820 1316 S 0.0 0.2 0:00.08 bash 28344 root 22 0 11932 1416 1184 S 0.0 0.1 0:00.00 mysqld_safe 18167 root 16 0 62628 1208 648 S 0.0 0.1 0:00.55 sshd 25941 root 15 0 12612 1192 928 R 0.0 0.1 0:02.21 top 11573 root 15 0 20876 1168 592 S 0.0 0.1 0:00.67 crond 9878 root 18 0 41000 1112 284 S 0.0 0.1 0:00.00 nginx 21736 root 23 0 21648 936 716 S 0.0 0.1 0:00.00 xinetd 11585 root 18 0 46748 816 428 S 0.0 0.1 0:00.00 saslauthd 14125 root 12 -4 12768 768 452 S 0.0 0.1 0:00.00 udevd 1 root 18 0 10352 728 616 S 0.0 0.1 0:17.93 init 24564 root 15 0 5912 680 544 S 0.0 0.1 0:01.90 syslogd 11618 root 18 0 46748 548 160 S 0.0 0.1 0:00.00 saslauthd Here's my php-fpm config: [global] pid = run/php-fpm.pid error_log = log/php-fpm.log log_level = notice [www] listen = 127.0.0.1:9000 user = www-data group = www-data pm = dynamic pm.max_children = 50 pm.start_servers = 3 pm.min_spare_servers = 3 pm.max_spare_servers = 10 pm.max_requests = 500 Sometimes it goes up to 400MB. And I'm planning to add a new website on my VPS. Is this normal?

  Read the article

• Single TCP-CLOSE_WAIT connection brings down application till restart

  - by broun

  Recently found a behaviour where my application had a connection in TCP Close_wait state till the app was restarted (after about 5 hours). But during this period the SUnreclaim space was also increasing constantly and went down on restart. The application is runnning on a rhel5 os and Im not very familiar with the memory management system. Would appreciate if someone clould tell me what extactly is the Ureclaim space and why it is increasing in sync with the close_wait.

  Read the article

• SUnreclaim size increasing in syc with a TCP-CLOSE_WAIT till application restart

  - by maver1k

  Recently found a behaviour where my application had a connection in TCP Close_wait state till the app was restarted (after about 5 hours). But during this period the SUnreclaim space was also increasing constantly and went down on restart. The application is runnning on a rhel5 os and Im not very familiar with the memory management system. Would appreciate if someone clould tell me what extactly is the Ureclaim space and why it is increasing in sync with the close_wait. Thanks.

  Read the article

• How can I limit Safari's RAM use?

  - by timothymh

  Safari uses a lot of RAM. Especially when you add Safari's use to WebProcess's use: it comes to about 1.5 GB of memory (for me, anyway). How can I limit the amount of RAM Safari and WebProcess use? I know a similar question was asked last summer, but two of those answers have to do with the program in question (Dropbox) and the other answer encourages use of Terminal, which I find terribly confusing. Are there any other methods?

  Read the article

• Ubuntu 14.04 says insufficient memory in my /boot memory alocation while updating

  - by Aravind Dollar

  I am new to Linux platform. I just tried installing Ubuntu alongside windows but allocated only 200 mb to the /boot partition which is not recommended. Now Ubuntu software update keep on insists me no enough space. What should I do? Is there any way to increase my /boot partition without removing the total OS? Or should I completely uninstall Ubuntu and put it back? Please anyone suggest me over this issue. I have no idea what to do. And please be elaborate, as I am new to Linux environment.

  Read the article

• strange memory usage pattern on windows server 2008 on login through remote desktop..

  - by headsling

  I'm running Windows Server 2008 Datacenter Service Pack 2 on a VM Ware instance with 10Gb ram allocated. I'm not running IIS or SQL Server. Under 'normal' conditions, the machine uses ~5.5Gb of memory. However, when I login to the server through remote desktop, the memory usage slowly climbs up to 9.8Gb of memory in use. After several minutes the memory slowly creeps back down to the 5.5Gb mark. I've tried killing all the processes associated with my login, on login, barring the taskmanager without success, and I can't see any process that is growing in memory usage when the memory is increasing. I'm assuming this is some system level cache that is growing / shrinking... but why is it doing this?

  Read the article

• Webcast - Oracle Database In-Memory Option

  - by Thanos Terentes Printzios

  Next to the recent announcement by Larry Ellison on the Future of the Database, we are happy to share this exclusive series of live webcasts from Oracle Database Product Management, where you can learn more about the brand new Oracle Database 12c In-Memory option. Oracle Database In-Memory is Oracle’s new memory-optimized technology that transparently accelerates analytic, data warehousing, and reporting workloads, while also accelerating transaction processing (OLTP) workloads. Participants will learn about Oracle Database In-Memory benefits, features, and leading edge architecture.  The Database In-Memory architecture provides the ability to easily process data orders of magnitude faster by simply enabling the feature and identifying tables to bring in-memory without application changes. Details on Oracle Database In-Memory’s ease of use and management, scalability, and availability will also be covered. Please join us to learn more about Oracle Database In-Memory and get first-hand knowledge of this important new feature. Delivery Format This FREE online LIVE eSeminar will be delivered over the Web.These Oracle webcasts are FREE for Customers, System Integrators, ISVs, VARs and Platform Partners. Presenter: Richard Jacobs, Oracle Solution Architect  Europe Webcast 1 Date: August 29, 2014 @ 10:00 am to 11:00 am Central European Summer Time (CEST)Register Here! Europe Webcast 2 Date: September 29, 2014 @ 10:00 am to 11:00 am Central European Summer Time (CEST)Register Here!

  Read the article

• PHP website Optimization

  - by ana

  I have a high traffic website and I need make sure my site is fast enough to display my pages to everyone rapidly. I searched on Google many articles about speed and optimization and here's what I found: Cache the page Save it to the disk Caching the page in memory: This is very fast but if I need to change the content of my page I have to remove it from cache and then re-save the file on the disk. Save it to disk This is very easy to maintain but every time the page is accessed I have to read on the disk. Which method should I go with? Thanks

  Read the article

• Performance optimization strategies of last resort?

  - by jerryjvl

  There are plenty of performance questions on this site already, but it occurs to me that almost all are very problem-specific and fairly narrow. And almost all repeat the advice to avoid premature optimization. Let's assume: the code already is working correctly the algorithms chosen are already optimal for the circumstances of the problem the code has been measured, and the offending routines have been isolated all attempts to optimize will also be measured to ensure they do not make matters worse What I am looking for here is strategies and tricks to squeeze out up to the last few percent in a critical algorithm when there is nothing else left to do but whatever it takes. Ideally, try to make answers language agnostic, and indicate any down-sides to the suggested strategies where applicable. I'll add a reply with my own initial suggestions, and look forward to whatever else the SO community can think of.

  Read the article

• Does MATLAB perform tail call optimization?

  - by Shea Levy

  I've recently learned Haskell, and am trying to carry the pure functional style over to my other code when possible. An important aspect of this is treating all variables as immutable, i.e. constants. In order to do so, many computations that would be implemented using loops in an imperative style have to be performed using recursion, which typically incurs a memory penalty due to the allocation a new stack frame for each function call. In the special case of a tail call (where the return value of a called function is immediately returned to the callee's caller), however, this penalty can be bypassed by a process called tail call optimization (in one method, this can be done by essentially replacing a call with a jmp after setting up the stack properly). Does MATLAB perform TCO by default, or is there a way to tell it to?

  Read the article

• Including associations optimization in Rails

  - by Vitaly

  Hey, I'm looking for help with Ruby optimization regarding loading of associations on demand. This is simplified example. I have 3 models: Post, Comment, User. References are: Post has many comments and Comment has reference to User (:author). Now when I go to the post page, I expect to see post body + all comments (and their respective authors names). This requires following 2 queries: select * from Post -- to get post data (1 row) select * from Comment inner join User -- to get comment + usernames (N rows) In the code I have: Post.find(params[:id], :include => { :comments => [:author] } But it doesn't work as expected: as I see in the back end, there're still N+1 hits (some of them are cached though). How can I optimize that?

  Read the article

• why optimization does not happen?

  - by aaa

  hi. I have C/C++ code, that looks like this: static int function(double *I) { int n = 0; // more instructions, loops, for (int i; ...; ++i) n += fabs(I[i] > tolerance); return n; } function(I); // return value is not used. compiler inlines function, however it does not optimize out n manipulations. I would expect compiler is able to recognize that value is never used as rhs only. Is there some side effect, which prevents optimization? Thanks

  Read the article

• Static variable for optimization

  - by keithjgrant

  I'm wondering if I can use a static variable for optimization: public function Bar() { static $i = moderatelyExpensiveFunctionCall(); if ($i) { return something(); } else { return somethingElse(); } } I know that once $i is initialized, it won't be changed by by that line of code on successive calls to Bar(). I assume this means that moderatelyExpensiveFunctionCall() won't be evaluated every time I call, but I'd like to know for certain. Once PHP sees a static variable that has been initialized, does it skip over that line of code? In other words, is this going to optimize my execution time if I make a lot of calls to Bar(), or am I wasting my time?

  Read the article

• Copy method optimization in compilers

  - by Dženan

  Hi All! I have the following code: void Stack::operator =(Stack &rhs) { //do the actual copying } Stack::Stack(Stack &rhs) //copy-constructor { top=NULL; //initialize this as an empty stack (which it is) *this=rhs; //invoke assignment operator } Stack& Stack::CopyStack() { return *this; //this statement will invoke copy contructor } It is being used like this: unsigned Stack::count() { unsigned c=0; Stack copy=CopyStack(); while (!copy.empty()) { copy.pop(); c++; } return c; } Removing reference symbol from declaration of CopyStack (returning a copy instead of reference) makes no difference in visual studio 2008 (with respect to number of times copying is invoked). I guess it gets optimized away - normally it should first make a copy for the return value, then call assignment operator once more to assign it to variable sc. What is your experience with this sort of optimization in different compilers? Regards, Dženan

  Read the article

• Common optimization rules

  - by mafutrct

  This is a dangerous question, so let me try to phrase it correctly. Premature optimization is the root of all evil, but if you know you need it, there is a basic set of rules that should be considered. This set is what I'm wondering about. For instance, imagine you got a list of a few thousand items. How do you look up an item with a specific, unique ID? Of course, you simply use a Dictionary to map the ID to the item. And if you know that there is a setting stored in a database that is required all the time, you simply cache it instead of issuing a database request hundred times a second. I guess there are a few even more basic ideas. I am specifically not looking for "don't do it, for experts: don't do it yet" or "use a profiler" answers, but for really simple, general hints. If you feel this is an argumentative question, you probably misunderstood my intention.

  Read the article

• Which isometric angles can be mirrored (and otherwise transformed) for optimization?

  - by Tom

  I am working on a basic isometric game, and am struggling to find the correct mirrors. Mirror can be any form of transform. I have managed to get SE out of SW, by scaling the sprite on X axis by -1. Same applies for NE angle. Something is bugging me, that I should be able to also mirror N to S, but I cannot manage to pull this one off. Am I just too sleepy and trying to do the impossible, or a basic -1 scale on Y axis is not enough? What are the common used mirror table for optimizing 8 angle (N, NE, E, SE, S, SW, W, NW) isometric sprites?

  Read the article

• When should an API favour optimization over readability and ease-of-use?

  - by jmlane

  I am in the process of designing a small library, where one of my design goals is to use as much of the native domain language as possible in the API. While doing so, I've noticed that there are some cases in the API outline where a more intuitive, readable attribute/method call requires some functionally unnecessary encapsulation. Since the final product will not necessarily require high performance, I am unconcerned about making the decision to favour ease-of-use in my current project over the most efficient implementation of the code in question. I know not to assume readability and ease-of-use are paramount in all expected use-cases, such as when performance is required. I would like to know if there are more general reasons that argue for an API design preferring (marginally) more efficient implementations?

  Read the article

• When should code favour optimization over readability and ease-of-use?

  - by jmlane

  I am in the process of designing a small library, where one of my design goals is that the API should be as close to the domain language as possible. While working on the design, I've noticed that there are some cases in the code where a more intuitive, readable attribute/method call requires some functionally unnecessary encapsulation. Since the final product will not necessarily require high performance, I am unconcerned about making the decision to favour ease-of-use in my current project over the most efficient implementation of the code in question. I know not to assume readability and ease-of-use are paramount in all expected use-cases, such as when performance is required. I would like to know if there are more general reasons that argue for a design preferring more efficient implementations—even if only marginally so?

  Read the article

• Segment register, IP register and memory addressing issue!

  - by Zia ur Rahman

  In the following text I asked two questions and I also described that what I know about these question so that you can understand my thinking. Your precious comments about the below text are required. Below is the Detail of 1ST Question As we know that if we have one mega byte memory then we need 20 bits to address this memory. Another thing is each memory cell has a physical address which is of 20 bits in 1Mb memory. IP register in IAPX88 is of 16 bits. Now my point of view is, we can not access the memory at all by the IP register because the memory need 20 bit address to be addressed but the IP register is of 16 bits. If we have a memory of 64k then IP register can access this memory because this memory needs 16 bits to be addressed. But incase of 1mb memory IP can’t.tell me am i right or not if not why? Suppose physical address of memory is 11000000000000000101 Now how can we access this memory location by 16 bits. Below is the detail of Next Question: My next question is , suppose IP register is pointing to memory location, and the segment register is also pointing to a memory location (start of the segment), the memory is of 1MB, how we can access a memory location by these two 16 bit registers tell me the sequence of steps how the 20 bits addressable memory location is accessed . If your answer is, we take the segment value and we shift it left by 4 bits and then add the IP value into it to get the 20 bits address, then this raises another question that is the address bus (the address bus should be 20 bits wide), the registers both the segment register and the IP register are of 16 bits each , now if address bus is 20 bits wide then this means that the address bus is connected to both these registers. If its not the case then another thing that comes into my mind is that both these registers generate a 20 bit address and there would be a register which can store 20 bits and this register would be connected to both these register and the address bus as well.

  Read the article

• Having problems with high CPU usage and apparent memory leak of Exim

  - by Dancrumb

  I'm having problems with my server and am hoping you can help. The culprit appears to be exim. The CPU usage is consistently high and the memory usage trends up and up and up for no apparent reason (this is not a heavily used server). To demonstrate the issue, I ran the following: root@server [/var/log]# service exim restart; for iter in `seq 0 9`; do date; top -n1 | grep exim; sleep 10; done Shutting down exim: [ OK ] Shutting down spamd: [ OK ] Starting exim: [ OK ] Sun Jun 6 18:12:07 CDT 2010 62592 root 25 0 11400 6572 2356 R 51.5 1.3 0:00.92 exim 62587 mailnull 18 0 7548 1212 792 S 0.0 0.2 0:00.00 exim Sun Jun 6 18:12:18 CDT 2010 62592 root 25 0 28768 23m 2356 R 57.4 4.6 0:06.75 exim 62587 mailnull 18 0 7548 1212 792 S 0.0 0.2 0:00.00 exim 62588 root 18 0 7536 2052 1648 S 0.0 0.4 0:00.00 exim Sun Jun 6 18:12:28 CDT 2010 62592 root 25 0 36408 30m 2356 R 55.5 6.0 0:12.59 exim 62587 mailnull 18 0 7548 1212 792 S 0.0 0.2 0:00.00 exim 62588 root 18 0 7536 2052 1648 S 0.0 0.4 0:00.00 exim Sun Jun 6 18:12:39 CDT 2010 62592 root 25 0 41396 35m 2356 R 53.5 7.0 0:18.35 exim 62587 mailnull 18 0 7548 1212 792 S 0.0 0.2 0:00.00 exim 62588 root 18 0 7536 2052 1648 S 0.0 0.4 0:00.00 exim Sun Jun 6 18:12:49 CDT 2010 62592 root 25 0 45868 40m 2356 R 47.5 7.8 0:24.06 exim 62587 mailnull 18 0 7548 1212 792 S 0.0 0.2 0:00.00 exim 62588 root 18 0 7536 2052 1648 S 0.0 0.4 0:00.00 exim Sun Jun 6 18:13:00 CDT 2010 62592 root 25 0 50056 44m 2356 R 55.3 8.6 0:29.84 exim 62587 mailnull 18 0 7548 1212 792 S 0.0 0.2 0:00.00 exim 62588 root 18 0 7536 2052 1648 S 0.0 0.4 0:00.00 exim Sun Jun 6 18:13:10 CDT 2010 62592 root 25 0 53888 47m 2356 R 55.2 9.4 0:35.63 exim 62587 mailnull 18 0 7548 1212 792 S 0.0 0.2 0:00.00 exim 62588 root 18 0 7536 2052 1648 S 0.0 0.4 0:00.00 exim Sun Jun 6 18:13:21 CDT 2010 62592 root 20 0 56920 50m 2356 R 55.3 9.9 0:41.15 exim 62587 mailnull 18 0 7548 1212 792 S 0.0 0.2 0:00.00 exim 62588 root 18 0 7536 2052 1648 S 0.0 0.4 0:00.00 exim Sun Jun 6 18:13:31 CDT 2010 62592 root 25 0 60380 54m 2356 R 53.4 10.6 0:46.98 exim 62587 mailnull 18 0 7548 1212 792 S 0.0 0.2 0:00.00 exim 62588 root 18 0 7536 2052 1648 S 0.0 0.4 0:00.00 exim Sun Jun 6 18:13:42 CDT 2010 62592 root 22 0 63400 57m 2356 R 49.5 11.2 0:52.74 exim 62587 mailnull 18 0 7548 1212 792 S 0.0 0.2 0:00.00 exim 62588 root 18 0 7536 2052 1648 S 0.0 0.4 0:00.00 exim After some time, it gets to a rate of picking up an extra MB every 10s. I've checked the exim logs and there are no messages coming in there. exim -bV shows: Exim version 4.69 #1 built 16-Mar-2009 14:44:43 Copyright (c) University of Cambridge 2006 Berkeley DB: Sleepycat Software: Berkeley DB 4.2.52: (February 22, 2005) Support for: crypteq iconv() IPv6 PAM Perl OpenSSL Content_Scanning Old_Demime Experimental_SPF Experimental_SRS Experimental_DomainKeys Lookups: lsearch wildlsearch nwildlsearch iplsearch dbm dbmnz passwd Authenticators: cram_md5 dovecot plaintext spa Routers: accept dnslookup ipliteral manualroute queryprogram redirect Transports: appendfile/maildir autoreply pipe smtp Size of off_t: 8 Configuration file is /etc/exim.conf I'm at something of a loss as to how to proceed. Any recommendations would be well received!

  Read the article

• C# WPF application is using too much memory while GC.GetTotalMemory() is low

  - by Dmitry

  I wrote little WPF application with 2 threads - main thread is GUI thread and another thread is worker. App has one WPF form with some controls. There is a button, allowing to select directory. After selecting directory, application scans for .jpg files in that directory and checks if their thumbnails are in hashtable. if they are, it does nothing. else it's adding their full filenames to queue for worker. Worker is taking filenames from this queue, loading JPEG images (using WPF's JpegBitmapDecoder and BitmapFrame), making thumbnails of them (using WPF's TransformedBitmap) and adding them to hashtable. Everything works fine, except memory consumption by this application when making thumbnails for big images (like 5000x5000 pixels). I've added textboxes on my form to show memory consumption (GC.GetTotalMemory() and Process.GetCurrentProcess().PrivateMemorySize64) and was very surprised, cuz GC.GetTotalMemory() stays close to 1-2 Mbytes, while private memory size constantly grows, especially when loading new image (~ +100Mb per image). Even after loading all images, making thumbnails of them and freeing original images, private memory size stays at ~700-800Mbytes. My VirtualBox is limited to 512Mb of physical memory and Windows in VirtualBox starts to swap alot to handle this huge memory consumption. I guess I'm doing something wrong, but I don't know how to investigate this problem, cuz according to GC, allocated memory size is very low. Attaching code of thumbnail loader class: class ThumbnailLoader { Hashtable thumbnails; Queue<string> taskqueue; EventWaitHandle wh; Thread[] workers; bool stop; object locker; int width, height, processed, added; public ThumbnailLoader() { int workercount,i; wh = new AutoResetEvent(false); thumbnails = new Hashtable(); taskqueue = new Queue<string>(); stop = false; locker = new object(); width = height = 64; processed = added = 0; workercount = Environment.ProcessorCount; workers=new Thread[workercount]; for (i = 0; i < workercount; i++) { workers[i] = new Thread(Worker); workers[i].IsBackground = true; workers[i].Priority = ThreadPriority.Highest; workers[i].Start(); } } public void SetThumbnailSize(int twidth, int theight) { width = twidth; height = theight; if (thumbnails.Count!=0) AddTask("#resethash"); } public void GetProgress(out int Added, out int Processed) { Added = added; Processed = processed; } private void AddTask(string filename) { lock(locker) { taskqueue.Enqueue(filename); wh.Set(); added++; } } private string NextTask() { lock(locker) { if (taskqueue.Count == 0) return null; else { processed++; return taskqueue.Dequeue(); } } } public static string FileNameToHash(string s) { return FormsAuthentication.HashPasswordForStoringInConfigFile(s, "MD5"); } public bool GetThumbnail(string filename,out BitmapFrame thumbnail) { string hash; hash = FileNameToHash(filename); if (thumbnails.ContainsKey(hash)) { thumbnail=(BitmapFrame)thumbnails[hash]; return true; } AddTask(filename); thumbnail = null; return false; } private BitmapFrame LoadThumbnail(string filename) { FileStream fs; JpegBitmapDecoder bd; BitmapFrame oldbf, bf; TransformedBitmap tb; double scale, dx, dy; fs = new FileStream(filename, FileMode.Open); bd = new JpegBitmapDecoder(fs, BitmapCreateOptions.None, BitmapCacheOption.OnLoad); oldbf = bd.Frames[0]; dx = (double)oldbf.Width / width; dy = (double)oldbf.Height / height; if (dx > dy) scale = 1 / dx; else scale = 1 / dy; tb = new TransformedBitmap(oldbf, new ScaleTransform(scale, scale)); bf = BitmapFrame.Create(tb); fs.Close(); oldbf = null; bd = null; GC.Collect(); return bf; } public void Dispose() { lock(locker) { stop = true; } AddTask(null); foreach (Thread worker in workers) { worker.Join(); } wh.Close(); } private void Worker() { string curtask,hash; while (!stop) { curtask = NextTask(); if (curtask == null) wh.WaitOne(); else { if (curtask == "#resethash") thumbnails.Clear(); else { hash = FileNameToHash(curtask); try { thumbnails[hash] = LoadThumbnail(curtask); } catch { thumbnails[hash] = null; } } } } } }

  Read the article

• Access cost of dynamically created objects with dynamically allocated members

  - by user343547

  I'm building an application which will have dynamic allocated objects of type A each with a dynamically allocated member (v) similar to the below class class A { int a; int b; int* v; }; where: The memory for v will be allocated in the constructor. v will be allocated once when an object of type A is created and will never need to be resized. The size of v will vary across all instances of A. The application will potentially have a huge number of such objects and mostly need to stream a large number of these objects through the CPU but only need to perform very simple computations on the members variables. Could having v dynamically allocated could mean that an instance of A and its member v are not located together in memory? What tools and techniques can be used to test if this fragmentation is a performance bottleneck? If such fragmentation is a performance issue, are there any techniques that could allow A and v to allocated in a continuous region of memory? Or are there any techniques to aid memory access such as pre-fetching scheme? for example get an object of type A operate on the other member variables whilst pre-fetching v. If the size of v or an acceptable maximum size could be known at compile time would replacing v with a fixed sized array like int v[max_length] lead to better performance? The target platforms are standard desktop machines with x86/AMD64 processors, Windows or Linux OSes and compiled using either GCC or MSVC compilers.

  Read the article

• How to compare a memory bits in C++?

  - by Trunet

  Hi, I need help with a memory bit comparison function. I bought a LED Matrix here with 4 x HT1632C chips and I'm using it on my arduino mega2560. There're no code available for this chipset(it's not the same as HT1632) and I'm writing on my own. I have a plot function that get x,y coordinates and a color and that pixel turn on. Only this is working perfectly. But I need more performance on my display so I tried to make a shadowRam variable that is a "copy" of my device memory. Before I plot anything on display it checks on shadowRam to see if it's really necessary to change that pixel. When I enabled this(getShadowRam) on plot function my display has some, just SOME(like 3 or 4 on entire display) ghost pixels(pixels that is not supposed to be turned on). If I just comment the prev_color if's on my plot function it works perfectly. Also, I'm cleaning my shadowRam array setting all matrix to zero. variables: #define BLACK 0 #define GREEN 1 #define RED 2 #define ORANGE 3 #define CHIP_MAX 8 byte shadowRam[63][CHIP_MAX-1] = {0}; getShadowRam function: byte HT1632C::getShadowRam(byte x, byte y) { byte addr, bitval, nChip; if (x>=32) { nChip = 3 + x/16 + (y>7?2:0); } else { nChip = 1 + x/16 + (y>7?2:0); } bitval = 8>>(y&3); x = x % 16; y = y % 8; addr = (x<<1) + (y>>2); if ((shadowRam[addr][nChip-1] & bitval) && (shadowRam[addr+32][nChip-1] & bitval)) { return ORANGE; } else if (shadowRam[addr][nChip-1] & bitval) { return GREEN; } else if (shadowRam[addr+32][nChip-1] & bitval) { return RED; } else { return BLACK; } } plot function: void HT1632C::plot (int x, int y, int color) { if (x<0 || x>X_MAX || y<0 || y>Y_MAX) return; if (color != BLACK && color != GREEN && color != RED && color != ORANGE) return; char addr, bitval; byte nChip; byte prev_color = HT1632C::getShadowRam(x,y); bitval = 8>>(y&3); if (x>=32) { nChip = 3 + x/16 + (y>7?2:0); } else { nChip = 1 + x/16 + (y>7?2:0); } x = x % 16; y = y % 8; addr = (x<<1) + (y>>2); switch(color) { case BLACK: if (prev_color != BLACK) { // compare with memory to only set if pixel is other color // clear the bit in both planes; shadowRam[addr][nChip-1] &= ~bitval; HT1632C::sendData(nChip, addr, shadowRam[addr][nChip-1]); shadowRam[addr+32][nChip-1] &= ~bitval; HT1632C::sendData(nChip, addr+32, shadowRam[addr+32][nChip-1]); } break; case GREEN: if (prev_color != GREEN) { // compare with memory to only set if pixel is other color // set the bit in the green plane and clear the bit in the red plane; shadowRam[addr][nChip-1] |= bitval; HT1632C::sendData(nChip, addr, shadowRam[addr][nChip-1]); shadowRam[addr+32][nChip-1] &= ~bitval; HT1632C::sendData(nChip, addr+32, shadowRam[addr+32][nChip-1]); } break; case RED: if (prev_color != RED) { // compare with memory to only set if pixel is other color // clear the bit in green plane and set the bit in the red plane; shadowRam[addr][nChip-1] &= ~bitval; HT1632C::sendData(nChip, addr, shadowRam[addr][nChip-1]); shadowRam[addr+32][nChip-1] |= bitval; HT1632C::sendData(nChip, addr+32, shadowRam[addr+32][nChip-1]); } break; case ORANGE: if (prev_color != ORANGE) { // compare with memory to only set if pixel is other color // set the bit in both the green and red planes; shadowRam[addr][nChip-1] |= bitval; HT1632C::sendData(nChip, addr, shadowRam[addr][nChip-1]); shadowRam[addr+32][nChip-1] |= bitval; HT1632C::sendData(nChip, addr+32, shadowRam[addr+32][nChip-1]); } break; } } If helps: The datasheet of board I'm using. On page 7 has the memory mapping I'm using. Also, I have a video of display working.

  Read the article

• Lua metatable Objects cannot be purge from memory?

  - by Prometheus3k

  Hi there, I'm using a proprietary platform that reported memory usage in realtime on screen. I decided to use a Class.lua I found on http://lua-users.org/wiki/SimpleLuaClasses However, I noticed memory issues when purging object created by this using a simple Account class. Specifically, I would start with say 146k of memory used, create 1000 objects of a class that just holds an integer instance variable and store each object into a table. The memory used is now 300k I would then exit, iterating through the table and setting each element in the table to nil. But would never get back the 146k, usually after this I am left using 210k or something similar. If I run the load sequence again during the same session, it does not exceed 300k so it is not a memory leak. I have tried creating 1000 integers in a table and setting these to nil, which does give me back 146k. In addition I've tried a simpler class file (Account2.lua) that doesn't rely on a class.lua. This still incurs memory fragmentation but not as much as the one that uses Class.lua Can anybody explain what is going on here? How can I purge these objects and get back the memory? here is the code --------Class.lua------ -- class.lua -- Compatible with Lua 5.1 (not 5.0). --http://lua-users.org/wiki/SimpleLuaClasses function class(base,ctor) local c = {} -- a new class instance if not ctor and type(base) == 'function' then ctor = base base = nil elseif type(base) == 'table' then -- our new class is a shallow copy of the base class! for i,v in pairs(base) do c[i] = v end c._base = base end -- the class will be the metatable for all its objects, -- and they will look up their methods in it. c.__index = c -- expose a ctor which can be called by () local mt = {} mt.__call = function(class_tbl,...) local obj = {} setmetatable(obj,c) if ctor then ctor(obj,...) else -- make sure that any stuff from the base class is initialized! if base and base.init then base.init(obj,...) end end return obj end c.init = ctor c.instanceOf = function(self,klass) local m = getmetatable(self) while m do if m == klass then return true end m = m._base end return false end setmetatable(c,mt) return c end --------Account.lua------ --Import Class template require 'class' local classname = "Account" --Declare class Constructor Account = class(function(acc,balance) --Instance variables declared here. if(balance ~= nil)then acc.balance = balance else --default value acc.balance = 2097 end acc.classname = classname end) --------Account2.lua------ local account2 = {} account2.classname = "unnamed" account2.balance = 2097 -----------Constructor 1 do local metatable = { __index = account2; } function Account2() return setmetatable({}, metatable); end end --------Main.lua------ require 'Account' require 'Account2' MAX_OBJ = 5000; test_value = 1000; Obj_Table = {}; MODE_ACC0 = 0 --integers MODE_ACC1 = 1 --Account MODE_ACC2 = 2 --Account2 TEST_MODE = MODE_ACC0; Lua_mem = ""; print("##1) collectgarbage('count'): " .. collectgarbage('count')); function Load() for i=1, MAX_OBJ do if(TEST_MODE == MODE_ACC0 )then table.insert(Obj_Table, test_value); elseif(TEST_MODE == MODE_ACC1 )then table.insert(Obj_Table, Account(test_value)); --Account.lua elseif(TEST_MODE == MODE_ACC2 )then table.insert(Obj_Table, Account2()); --Account2.lua Obj_Table[i].balance = test_value; end end print("##2) collectgarbage('count'): " .. collectgarbage('count')); end function Purge() --metatable purge if(TEST_MODE ~= MODE_ACC0)then --purge stage 0: print("set each elements metatable to nil") for i=1, MAX_OBJ do setmetatable(Obj_Table[i], nil); end end --purge stage 1: print("set table element to nil") for i=1, MAX_OBJ do Obj_Table[i] = nil; end --purge stage 2: print("start table.remove..."); for i=1, MAX_OBJ do table.remove(Obj_Table, i); end print("...end table.remove"); --purge stage 3: print("create new object_table {}"); Obj_Table= {}; --purge stage 4: print("collectgarbage('collect')"); collectgarbage('collect'); print("##3) collectgarbage('count'): " .. collectgarbage('count')); end --Loop callback function OnUpdate() collectgarbage('collect'); Lua_mem = collectgarbage('count'); end ------------------- --NOTE: --On start of game runs Load(), another runs Purge() --Update I've updated the code with suggestions from comments below, and will post my findings later today.

  Read the article

• Can knowing C actually hurt the code you write in higher level languages?

  - by Jurily

  The question seems settled, beaten to death even. Smart people have said smart things on the subject. To be a really good programmer, you need to know C. Or do you? I was enlightened twice this week. The first one made me realize that my assumptions don't go further than my knowledge behind them, and given the complexity of software running on my machine, that's almost non-existent. But what really drove it home was this Slashdot comment: The end result is that I notice the many naive ways in which traditional C "bare metal" programmers assume that higher level languages are implemented. They make bad "optimization" decisions in projects they influence, because they have no idea how a compiler works or how different a good runtime system may be from the naive macro-assembler model they understand. Then it hit me: C is just one more abstraction, like all others. Even the CPU itself is only an abstraction! I've just never seen it break, because I don't have the tools to measure it. I'm confused. Has my mind been mutilated beyond recovery, like Dijkstra said about BASIC? Am I living in a constant state of premature optimization? Is there hope for me, now that I realized I know nothing about anything? Is there anything to know, even? And why is it so fascinating, that everything I've written in the last five years might have been fundamentally wrong? To sum it up: is there any value in knowing more than the API docs tell me? EDIT: Made CW. Of course this also means now you must post examples of the interpreter/runtime optimizing better than we do :)

  Read the article

< Previous Page | 29 30 31 32 33 34 35 36 37 38 39 40  | Next Page >