Search Results

Search found 13151 results on 527 pages for 'performance counters'.

Page 68/527 | < Previous Page | 64 65 66 67 68 69 70 71 72 73 74 75  | Next Page >

• Quantifying the effects of partition mis-alignment

  - by Matt

  I'm experiencing some significant performance issues on an NFS server. I've been reading up a bit on partition alignment, and I think I have my partitions mis-aligned. I can't find anything that tells me how to actually quantify the effects of mis-aligned partitions. Some of the general information I found suggests the performance penalty can be quite high (upwards of 60%) and others say it's negligible. What I want to do is determine if partition alignment is a factor in this server's performance problems or not; and if so, to what degree? So I'll put my info out here, and hopefully the community can confirm if my partitions are indeed mis-aligned, and if so, help me put a number to what the performance cost is. Server is a Dell R510 with dual E5620 CPUs and 8 GB RAM. There are eight 15k 2.5” 600 GB drives (Seagate ST3600057SS) configured in hardware RAID-6 with a single hot spare. RAID controller is a Dell PERC H700 w/512MB cache (Linux sees this as a LSI MegaSAS 9260). OS is CentOS 5.6, home directory partition is ext3, with options “rw,data=journal,usrquota”. I have the HW RAID configured to present two virtual disks to the OS: /dev/sda for the OS (boot, root and swap partitions), and /dev/sdb for a big NFS share: [root@lnxutil1 ~]# parted -s /dev/sda unit s print Model: DELL PERC H700 (scsi) Disk /dev/sda: 134217599s Sector size (logical/physical): 512B/512B Partition Table: msdos Number Start End Size Type File system Flags 1 63s 465884s 465822s primary ext2 boot 2 465885s 134207009s 133741125s primary lvm [root@lnxutil1 ~]# parted -s /dev/sdb unit s print Model: DELL PERC H700 (scsi) Disk /dev/sdb: 5720768639s Sector size (logical/physical): 512B/512B Partition Table: gpt Number Start End Size File system Name Flags 1 34s 5720768606s 5720768573s lvm Edit 1 Using the cfq IO scheduler (default for CentOS 5.6): # cat /sys/block/sd{a,b}/queue/scheduler noop anticipatory deadline [cfq] noop anticipatory deadline [cfq] Chunk size is the same as strip size, right? If so, then 64kB: # /opt/MegaCli -LDInfo -Lall -aALL -NoLog Adapter #0 Number of Virtual Disks: 2 Virtual Disk: 0 (target id: 0) Name:os RAID Level: Primary-6, Secondary-0, RAID Level Qualifier-3 Size:65535MB State: Optimal Stripe Size: 64kB Number Of Drives:7 Span Depth:1 Default Cache Policy: WriteBack, ReadAdaptive, Direct, No Write Cache if Bad BBU Current Cache Policy: WriteThrough, ReadAdaptive, Direct, No Write Cache if Bad BBU Access Policy: Read/Write Disk Cache Policy: Disk's Default Number of Spans: 1 Span: 0 - Number of PDs: 7 ... physical disk info removed for brevity ... Virtual Disk: 1 (target id: 1) Name:share RAID Level: Primary-6, Secondary-0, RAID Level Qualifier-3 Size:2793344MB State: Optimal Stripe Size: 64kB Number Of Drives:7 Span Depth:1 Default Cache Policy: WriteBack, ReadAdaptive, Direct, No Write Cache if Bad BBU Current Cache Policy: WriteThrough, ReadAdaptive, Direct, No Write Cache if Bad BBU Access Policy: Read/Write Disk Cache Policy: Disk's Default Number of Spans: 1 Span: 0 - Number of PDs: 7 If it's not obvious, virtual disk 0 corresponds to /dev/sda, for the OS; virtual disk 1 is /dev/sdb (the exported home directory tree).

  Read the article

• Request bursting from web application Load Tests

  - by MaseBase

  I'm migrating our web and database hosting to a new environment on all new machines. I've recently performed a Load Test using WAPT to generate load from multiple distributed clients. The server has plenty of room to handle the traffic load, but I'm seeing an odd pattern of incoming traffic during the load tests. Here is the gist of our setup: Firewall server running MS Forefront TMG 2010 on Win 2k8 server Request routing done by IIS Application Request Routing on firewall machine Web server is a Hyper-V VM on the Database server (which is the host OS) These machines are hefty with dual-CPU's with six cores (12 total procs) Web server running IIS 7.5 Web applications built in ASP.NET 2.0, with 1 ISAPI filter (Url Rewrite) in front What I'm seeing during the load tests is that the requests all come through in bursts. Even though I have 7 different distributed clients sending traffic loads, the requests come through about 300-500 requests at a time. The performance monitor shows nearly all of the counters moving through this pattern, where a burst of requests comes in the req/sec jumps to 70, the queued requests jumps to 500, the current requests jumps up, the CPU jumps up, everything. Then once it's handled that group of requests, it has a lull for nearly 10 seconds where nearly nothing is happening. 0-5 req/sec, 0 queued requests, minimal CPU usage. Then after 10 seconds of inactivity, another burst comes through, spiking all of the counters once again. What I can't figure out is why the requests are coming through in bursts when I know that the load being generated is not sent that way, especially considering the various load-generating clients sending traffic all in different intervals with random think time's between each request. Is there something in the layers between Hyper-V or perhaps in the hardware which might cause this coalesce of requests together? Here is what i'm looking at, the highlighted metric is Requests/sec, but the others critical counter go with it: Requests Queued (which I'd obviously like to keep as close to 0 as possible). Any ideas on this?

  Read the article

• How to test TempDB performance?

  - by Matt Penner

  I'm getting some conflicting advice on how to best configure our SQL storage with our current SAN. I would like to do some of my own performance testing with a few different configurations. I looked at using SQLIOSim but it doesn't seem to simulate TempDB. Can anyone recommend a way to test data, log and TempDB performance? What about using a SQL profiler trace file from our production system? How would I use This to run against my test server? Thanks, Matt

  Read the article

• What are the performance differences between PCI-Express x16 and x4

  - by Cestarian

  I have two PCI-Express 2.0 x16 slots on my motherboard, but one of them is actually just x4. For passing through my graphics card to a virtual machine I need to use both slots simultaneously and unfortunately, the easiest way for me to achieve that is putting the stronger card in the x4 slot (secondary slot; I need the stronger card to be in the secondary slot, not the primary). As such I am wondering what sort of noticable performance differences I can expect from using the x4 slot with a strong card as opposed to having it in the true x16 slot. Does it limit the performance so much that the strong card in the x4 slot will actually perform worse than the significantly weaker card in the x16 slot? (For spec comparison I am using a GTX-670 in the x4 slot and GTX-550-Ti in the x16 slot) What implications does this have?

  Read the article

• Using SSD as disk cache

  - by casualcoder

  Is there software for Linux to use an SSD as disk cache? I believe that Sun does something like this with ZFS, though not sure. A quick search provides nothing suitable. The goal would be to put frequently requested files on the SSD on-the-fly. Since the SSD has more capacity than RAM for less money and better performance than hard disk, this should provide an efficient performance boost.

  Read the article

• Zero-channel RAID for High Performance MySQL Server (IBM ServeRAID 8k) : Any Experience/Recommendati

  - by prs563

  We are getting this IBM rack mount server and it has this IBM ServeRAID8k storage controller with Zero-Channel RAID and 256MB battery backed cache. It can support RAID 10 which we need for our high performance MySQL server which will have 4 x 15000K RPM 300GB SAS HDD. This is mission-critical and we want as much bandwidth and performance. Is this a good card or should we replace with another IBM RAID card? IBM ServeRAID 8k SAS Controller option provides 256 MB of battery backed 533 MHz DDR2 standard power memory in a fixed mounting arrangement. The device attaches directly to IBM planar which can provide full RAID capability. Manufacturer IBM Manufacturer Part # 25R8064 Cost Central Item # 10025907 Product Description IBM ServeRAID 8k SAS - Storage controller (zero-channel RAID) - RAID 0, 1, 5, 6, 10, 1E Device Type Storage controller (zero-channel RAID) - plug-in module Buffer Size 256 MB Supported Devices Disk array (RAID) Max Storage Devices Qty 8 RAID Level RAID 0, RAID 1, RAID 5, RAID 6, RAID 10, RAID 1E Manufacturer Warranty 1 year warranty

  Read the article

• optimizing file share performance on Win2k8?

  - by Kirk Marple

  We have a case where we're accessing a RAID array (drive E:) on a Windows Server 2008 SP2 x86 box. (Recently installed, nothing other than SQL Server 2005 on the server.) In one scenario, when directly accessing it (E:\folder\file.xxx) we get 45MBps throughput to a video file. If we access the same file on the same array, but through UNC path (\server\folder\file.xxx) we get about 23MBps throughput with the exact same test. Obviously the second test is going through more layers of the stack, but that's a major performance hit. What tuning should we be looking at for making the UNC path be closer in performance to the direct access case? Thanks, Kirk (corrected: it is CIFS not SMB, but generalized title to 'file share'.) (additional info: this happens during the read from a single file, not an issue across multiple connections. the file is on the local machine, but exposed via file share. so client and file server are both same Windows 2008 server.)

  Read the article

• Defragment / Performance Monitor without Task Scheduler

  - by mjaggard

  My organisation has a policy of disabling Task Scheduler on all servers and workstations (don't ask, I tried once to wrestle the pig). I need to collect performance stats using Data Collector Sets in Windows 7 or Windows 2008 but the Performance Monitor interface requires Task Scheduler to be running. Is this possible because I'm not trying to schedule anything (except the collection of WMI information every 15 seconds but I doubt it hands that task off to the task scheduler)? Is there any way to trick it into thinking Task Scheduler is running? If not, is there any way to temporarily override the group policy to allow Task Scheduler to run? I've found that most group policy can be overridden in this way by an Administrator by editing the registry. On exactly the same vein, I want to defragment a hard disk on one of my workstations, but I can't get it to start because of the dependancy on Task Scheduler - is it possible to overcome this?

  Read the article

• Windows Server 2008 Core - performance

  - by Feryt

  Are there any performance advantages over classis windows server 2008 installation?

  Read the article

• Any reason not to disable the Windows pagefile given enough physical RAM?

  - by Evgeny

  The question of disabling the Windows pagefile has already been discussed quite a bit, for example here and here and here. People continue to upvote answers that say "you should not disable your pagefile even if you have plenty of RAM", but I have yet to see any concrete, verifiable reasons being given for this advice. As far as I can see, if you never need to read from the pagefile (because you have enough RAM) then performance could only be worse with it enabled due to Windows pre-emptively writing to it. At best, performance would be the same. I can't see how it could possibly be improved by writing data you never need to read. So my question is: Assuming that I have enough physical RAM for everything I do, is there any reason I should not disable the pagefile? Let's say the version of Windows is Windows XP x64 SP2 or Windows Server 2003 x64 SP2 (same thing). If it's different for Windows Server 2008 x64 I'd be interested to hear an answer for that as well. I'm looking for specific, objective reasons from good sources, not just opinions. Something like "here are the benchmarks done with and without a pagefile and the results were better with a pagefile, even with enough RAM" or "according to this MS KB article problem X occurs if you disable the pagefile". So far the only reasons I've seen mentioned are: Even if you think you have enough RAM you might run out. OK, but for the purposes of this question, let's just take it as a given that I have enough. Maybe I only ever read my email and I have 16GB RAM. Or 128GB. Or 1TB. Or whatever - but it's enough for 100% of what I do, 100% of the time. Another way to think of it is: if I have x MB physical RAM and y MB pagefile and I never run out of RAM in that configuration, would I not be better off, performance-wise, with x+y MB physical RAM and no pagefile? Windows is "used to" having a paging file and it might not function as reliably (from Understanding the Impact of RAM on Overall System Performance That's rather vague and I find it hard to believe, given that MS has provided the option to disable the pagefile. Windows knows what it's doing better than you. No - it doesn't know that I won't run more programs or load more data, but I do.

  Read the article

• Should I run my database off of a RAID 5 configuration?

  - by Scott Saad

  I've heard that the write performance of RAID 5 can be appalling at times. While I want the redundancy that it provides I don't want to sacrifice my database insert/update times. Is this something I should be worried about and if so, what would be the recommendation to get redundancy with good write performance?

  Read the article

• Peforming an Audit for SQL Server 2008

  - by Nai

  Hi all, Do you guys have any good step by step type links for performing an SQL Server 2008 Performance Audit? I know Brad McGehee has written extensively on this but for SQL Server 2005 over at http://www.sql-server-performance.com. But are any such articles for SQL Server 2008? Thanks!

  Read the article

• How can ASP.NET's "Request Wait Time" be 0 when "Requests Queued" is consistently in the hundreds?

  - by ondrej

  I'm curious why Performance Monitor claims I always have a few hundred ASP.NET 3.5 requests "queued". The "Requests Queued" "ASP.NET v2.0.50727" performance counter is hovering in the few-hundred range despite the fact "Request Wait Time" is consistently 0. If each and every request never waits even a fraction of a millisecond, how could it be in the queue? The "ASP.NET Apps v2.0.50727" counters for "Requests In Application Queue" and "Request Wait Time" are always 0.

  Read the article

• increase performance of virtual machines on the 27" imac

  - by evan

  I'm using a 27" iMac (i7, 8GB RAM) at work and normally run two or three virtual machines at the same time, which hurts the performance of each virtual machine. I've learned on these forums the best way to increase virtual machine performance (aside from RAM) is to have them running on a separate hard drive from the one the OS is on. Of course with the iMac you can only have one hard drive and not even an SAS or solid state drive (well you could probably take it apart and put one in yourself but I wouldn't be permitted to do that). That being said, do you think it would help to run one or more virtual machines from a firewire external drive (or a usb 2.0)? Thanks for your input!

  Read the article

• MD3200i Slow Performance and Queue Depth

  - by Caleb_S

  Read performance on our SAN is slow under certain workloads. When we compare this to some local storage, we find the local storage performing 2x as fast. The SAN performs well with a high Queue Depth, and poorly with a low queue depth. However, the local storage performs well with a low Queue Depth. I'd like to know the reason for this occurring and find out what the specific limiting factor is in this situation. MD3200i iSCSI SAN ($15,000) 6 x 600GB 15k SAS RAID5 6 x 2TB 7.2k NLS RAID5 XCOPY /j Benchmark: (Slow) 15k Array - 71MB/s (Queue Depth 1) 7.2k Array- 71MB/s (Queue Depth 1) Robycopy /MT:32 Benchmark: (Fast) 15k Array - 171MB/s (Queue Depth ~12) 7.2k Array- 128MB/s (Queue Depth ~12) , , Read Performance on a Local controller is fast under the workload the SAN is slow at. , HighPoint 2230 RAID Controller ($600) 4 x 1TB 7.2k SATA RAID5 XCOPY /j Benchmark: 7.2k Array - 145MB/s (Queue Depth 1) (appears to max out the SATA bus)

  Read the article

• Tools to monitor guest OS performance in vSphere

  - by Quick Joe Smith

  I am looking for some tool or way to retrieve performance data from guest VMs running under vSphere 4.1. I am currently interested in the 4 basic metrics: CPU(%), Memory(%), Disk availability(%) & Network utilisation(Kb/s). The issue I have is that all of vSphere's performance data is from a ESXi host perspective (active, shared, consumed, overhead, swapped etc.) which is far removed from the data from the VM's own perspective. For instance, I have a Windows server VM idling, using around 410MB (~25% of its allocated 2GB) as reported by Task Manager, and this is the value I'm after. vSphere's metrics seem unable to arrive at this figure by any reliable and repeatable means. Is anyone aware of tools that can obtain this kind of data? The simpler, the better.

  Read the article

• Cassandra on heterogeneous servers

  - by happy-coding

  I am currently running 4 cassandra nodes with the following hardware in a Apache Cassandra cluster: AMD Athlon 64 X2 6000+ 8G RAM 750G hard disk It shows not such a good writing performance and a really bad read performance with sometimes also timeouts. I was wondering if it makes sense to add 2 nodes with a different hardware (8 CPUs and more RAM) to improve this. Or does a cassandra cluster works best with the same hardware in every node? Thanks & best regards

  Read the article

• a load balancing scenario using HAProxy and keepalived shows no performance advantage

  - by chakoshi

  Hi, I am trying to setup a load balanced web server scenario, using two HAproxy load balancers and two debian web servers following this guide http://www.howtoforge.com/setting-up-a-high-availability-load-balancer-with-haproxy-keepalived-on-debian-lenny. the setup is working but the results of simple performance benchmarking is not what I expected. I tried apache benchmark tool to send lots of requests to servers (one time directly testing one of the web servers and the other time testing through the load balancer) using the command "ab -n 1000000 -c 500 http://IP/index.html", but the test results shows better performance for the single server without load balancer. can any one tell me if I'm going wrong on some thing?

  Read the article

• TCP/IP performance tuning under KVM/Qemu

  - by vpetersson

  With more and more companies switching to public cloud services, I'm curious what you guys' thoughts are on TCP/IP tuning in the cloud. Is it worth bothering with? Given that you don't have access to the host-server, you're somewhat limited I presume Let's say for the sake of the argument that you're running three MongoDB-servers in a replica-set on FreeBSD or Linux that all sync over an internal network. I'd also be curious if anyone made any actual performance benchmarks to back up their arguments. I benchmarked the various network drivers available for KVM/Qemu here, but I'm curious what the gurus here suggest to tune further. I started playing around a bit with the tuning-recommendations as suggested over here, but interestingly enough I saw a decrease in performance, rather than an increase, but perhaps I didn't fully understand the tweaks. Update: I did a few more benchmarks and posted the result here. Unfortunately the result wasn't really what I expected.

  Read the article

• iSCSI SAN RAID 10 Performance -- Poor Read, Good Write

  - by Litzner

  I have a EqualLogic PS4000 SAN unit with the latest firmware, setup in RAID 10. I have 3 2TB Volumes on the SAN shared out via iSCSI on 2 eth ports on two different subnets. I have moved a test server over to this newly setup SAN, and my testing is showing me a problem. I am getting dismal read performance in everything except a test with 32 queue depth (see attach image) Write performance seems to be right about where it should be. I have tried MPIO on and off, on was slightly better but not much.

  Read the article

• Addons which actually make Firefox run faster?

  - by Zombies

  I would like to know of addons which actually enhance firefox's performance, both intentionally and unintentionally. I find that firefox tends to have major performance issues with certain websites. These websites tend to have a fair amount of javascript and css, and probably a large dom tree which may even be growing dynamically through javascript too. The worse offenders are those with heavy javascript, use heavy facebook integration, websites with non performant javascript, excessive javascript and websites with too many advertisements.

  Read the article

• What is the best way to secure MySQL data on a laptop *without* whole-disk-encryption?

  - by GJ

  I need to have the mysql data on my laptop stored in an encrypted state so that in case of the laptop being lost/stolen it will extremely difficult to recover the data without the password. I don't wish to use whole disk encryption, due to the performance impact it will have on other disk-intensive programs' usage. What could be the ideal solution for me balancing security and performance? Thanks!

  Read the article

• which packet delay and db response time is the least and best?

  - by PacketDelay

  In terms of performance for delay time and response time which of the following would be the best in terms of performance? ie least delay and response time. 1) 0.4 2) 0.004 3) 0.000030 4) 0.000000020 I'm trying to interpret some graphs but these numbers sort of confuse me. I think they are milliseconds? Thank You

  Read the article

• Improving Windows Authentication performance on IIS

  - by flalar

  We're struggling with performance issues with a ASP.NET MVC site that is using Windows Authentication. Response time is very slow on the first request to the site when the user is being authenticated. Further, every time the Authorization header is sent from the browser the response time increases with many seconds. The same issue occurs for both executed files and static content like CSS and JS. Access to the application is restricted to users within a certain role and we are now planning to allow access to static files for all authenticated users to see if that helps. The authentication method in use is NTLM. How should we go forward in pinpointing why authentication decreases performance drastically?

  Read the article

• nginx: URL rewrites and performance

  - by j0nes

  I have a website where I need to change the URL structure. The old URLs look like /olddir/part1_de.htm, the new ones will look like /newdir/sub/category/anotherpage.htm. There are a lot of URL rewrites I need to do, I assume about 500 distinct rewrites in the end. As my website gets quite a lot of traffic, my main concern is about performance at the moment. My questions are: I assume that for each request, the rewrites block will be parsed and the regex will be evaluated. Am I right? Will there be a performance penalty if I use these rewrites? Can nginx handle this? Are there any "best practices" to follow when doing a lot of rewrites?

  Read the article

< Previous Page | 64 65 66 67 68 69 70 71 72 73 74 75  | Next Page >