Search Results

Search found 11380 results on 456 pages for 'cpu speed'.

Page 53/456 | < Previous Page | 49 50 51 52 53 54 55 56 57 58 59 60  | Next Page >

• Nginx and low-speed connections: request terminates after 253 seconds

  - by meze

  I'm trying to make nginx to handle static files. All is working fine except that when I throttle my connection speed to 8kbit/s, the loading process of a file just stops after 253-255 seconds (4.2 min according to chrome). No error in the log, the status code is 200 but the response is received partially. If I disable nginx and make apache to send the same file - it loads successfully after 10 minutes. The config I use for debugging is: client_header_buffer_size 16k; large_client_header_buffers 4 8k; client_max_body_size 50m; client_body_buffer_size 16k; client_header_timeout 20m; client_body_timeout 20m; send_timeout 20m; Did i miss some configurations?

  Read the article

• Mac OS X Server 10.6.6 "disables" CPU in VMware Fusion [closed]

  - by wjlafrance

  Hello! I installed Mac OS X Server 10.6.0 in VMWare Fusion the other night and it worked perfectly, until I ran Software Update. I upgraded to 10.6.6 through the combo updater, and now when I start the VM it says: "The CPU has been disabled by the guest operating system. You will need to power off or reset the virtual machine at this point." I've switched the operating system in the options to OS X Server 32bit, 64bit, and even to Windows 7, and nothing has worked. Does anyone have any ideas?

  Read the article

• Mysql process goes over 100% of CPU usage

  - by Temnovit

  Hello! I'm experiencing some problems with my LAMP server. Recently, everything became very slow, even though visitor count on my websites didn't change to much. When I run top command, it sais that mysql process has taken over 150-200% of CPU. How's that possible, I always thought that 100% is a maximum? I'm running Ubuntu 9.04 server edition with 1,5 GB RAM my.cnf settings: key_buffer = 64M max_allowed_packet = 16M thread_stack = 192K thread_cache_size = 8 myisam-recover = BACKUP max_connections = 200 table_cache = 512 table_definition_cache = 512 thread_concurrency = 2 read_buffer_size = 1M sort_buffer_size = 4M join_buffer_size = 1M query_cache_limit = 1M # the maximum size of individual query results query_cache_size = 128M Here is the output of MySQLTuner: The top command: What could be the cause of this problem? Can I make changes to my my.cnf to prevent server from hanging?

  Read the article

• Slow speed for UFS mounted drive in Linux

  - by Incredible

  Hi, I have a disk that has Sun OS disk, (ufs filesystem). And I want to mount it in my debain machine with read/write mode. Since by deafult linux doesn't support write to ufs filesystem. I had to recompile the kernel by setting to the flag CONFIG_UFS_FS_WRITE=y. Now I am able to write to the filesystem, but the read/write speed is very slow. It is around 120 KB/s. Any idea what is wrong and how to resolve this issue? Thank you in advance.

  Read the article

• Webserver - Memory-bound or CPU-bound? [closed]

  - by JJP

  Possible Duplicate: How do you do Load Testing and Capacity Planning for Web Sites I'm installing a social network using Zend Framework & MySql, with lots of plugins & queries. I want Webserver & Sql server on one box. I'm trying to choose between two machines (on hetzner.de): A) intel i7-2600 3.4 GHz 16 GB DDR3 RAM B) intel i7-920 2.6 GHz 24 GB DDR3 RAM B has 50% more RAM but 30% slower clock speed. Q is: is it obvious where the bottleneck will be? Would I ever need 24GB of RAM, even with lots of concurrent users?

  Read the article

• MSSQL 2005 Snapshot Agent Uses 100% CPU

  - by matth1jd

  When setting up a new subscription to a publication (transactional replication) from 64-bit SQL Server 2005 to 64-bit SQL Server 2005 the Snapshot Agent on the publisher consumes 100% of the CPU. I am using SSMS to create the new subscription. My initial impression is that this could be from row locking occurring during the generation of the snapshot but I have read that a concurrent snapshot is generated by default in SQL Server 2005, and that row locking shouldn't be a concern. As this is a production server I would like to be able to initialize replication without bringing the box to it's knees. Any suggestions would be helpful and appreciated.

  Read the article

• sqlserver.exe uses 100% CPU

  - by Markus

  I've created an application (asp.net) that once a day syncs an entire database through XML-files. The sync first creates an transaction and then clears the databases tables and then starts to parse and insert the new rows into the database. When all the parsing is complete it commits the transaction. This works fine on a SQL Server 2005 (on another machine), but on SQL Server 2005 Express, the process starts to use 100% CPU after a while, and as I log the inserts being made I can see that it just stops inserting. No exception, it just stops inserting. Anyone got any idea what this may be? I've previously run the syncronization on another sql 2005 express (also on another computer), and that worked. The server has only 2GB RAM, could this be the problem?

  Read the article

• A Patent for Workload Management Based on Service Level Objectives

  - by jsavit

  I'm very pleased to announce that after a tiny :-) wait of about 5 years, my patent application for a workload manager was finally approved. Background Many operating systems have a resource manager which lets you control machine resources. For example, Solaris provides controls for CPU with several options: shares for proportional CPU allocation. If you have twice as many shares as me, and we are competing for CPU, you'll get about twice as many CPU cycles), dedicated CPU allocation in which a number of CPUs are exclusively dedicated to an application's use. You can say that a zone or project "owns" 8 CPUs on a 32 CPU machine, for example. And, capped CPU in which you specify the upper bound, or cap, of how much CPU an application gets. For example, you can throttle an application to 0.125 of a CPU. (This isn't meant to be an exhaustive list of Solaris RM controls.) Workload management Useful as that is (and tragic that some other operating systems have little resource management and isolation, and frighten people into running only 1 app per OS instance - and wastefully size every server for the peak workload it might experience) that's not really workload management. With resource management one controls the resources, and hope that's enough to meet application service objectives. In fact, we hold resource distribution constant, see if that was good enough, and adjust resource distribution if that didn't meet service level objectives. Here's an example of what happens today: Let's try 30% dedicated CPU. Not enough? Let's try 80% Oh, that's too much, and we're achieving much better response time than the objective, but other workloads are starving. Let's back that off and try again. It's not the process I object to - it's that we to often do this manually. Worse, we sometimes identify and adjust the wrong resource and fiddle with that to no useful result. Back in my days as a customer managing large systems, one of my users would call me up to beg for a "CPU boost": Me: "it won't make any difference - there's plenty of spare CPU to be had, and your application is completely I/O bound." User: "Please do it anyway." Me: "oh, all right, but it won't do you any good." (I did, because he was a friend, but it didn't help.) Prior art There are some operating environments that take a stab about workload management (rather than resource management) but I find them lacking. I know of one that uses synthetic "service units" composed of the sum of CPU, I/O and memory allocations multiplied by weighting factors. A workload is set to make a target rate of service units consumed per second. But this seems to be missing a key point: what is the relationship between artificial 'service units' and actually meeting a throughput or response time objective? What if I get plenty of one of the components (so am getting enough service units), but not enough of the resource whose needed to remove the bottleneck? Actual workload management That's not really the answer either. What is needed is to specify a workload's service levels in terms of externally visible metrics that are meaningful to a business, such as response times or transactions per second, and have the workload manager figure out which resources are not being adequately provided, and then adjust it as needed. If an application is not meeting its service level objectives and the reason is that it's not getting enough CPU cycles, adjust its CPU resource accordingly. If the reason is that the application isn't getting enough RAM to keep its working set in memory, then adjust its RAM assignment appropriately so it stops swapping. Simple idea, but that's a task we keep dumping on system administrators. In other words - don't hold the number of CPU shares constant and watch the achievement of service level vary. Instead, hold the service level constant, and dynamically adjust the number of CPU shares (or amount of other resources like RAM or I/O bandwidth) in order to meet the objective. Instrumenting non-instrumented applications There's one little problem here: how do I measure application performance in a way relating to a service level. I don't want to do it based on internal resources like number of CPU seconds it received per minute - We need to make resource decisions based on externally visible and meaningful measures of performance, not synthetic items or internal resource counters. If I have a way of marking the beginning and end of a transaction, I can then measure whether or not the application is meeting an objective based on it. If I can observe the delay factors for an application, I can see which resource shortages are slowing an application enough to keep it from meeting its objectives. I can then adjust resource allocations to relieve those shortages. Fortunately, Solaris provides facilities for both marking application progress and determining what factors cause application latency. The Solaris DTrace facility let's me introspect on application behavior: in particular I can see events like "receive a web hit" and "respond to that web hit" so I can get transaction rate and response time. DTrace (and tools like prstat) let me see where latency is being added to an application, so I know which resource to adjust. Summary After a delay of a mere few years, I am the proud creator of a patent (advice to anyone interested in going through the process: don't hold your breath!). The fundamental idea is fairly simple: instead of holding resource constant and suffering variable levels of success meeting service level objectives, properly characterise the service level objective in meaningful terms, instrument the application to see if it's meeting the objective, and then have a workload manager change resource allocations to remove delays preventing service level attainment. I've done it by hand for a long time - I think that's what a computer should do for me.

  Read the article

• Collision point of 2 curves in a 3d-room

  - by Frank

  Hello, i am programming a small game for quite some time. We started coding a small FPS-Shooter inside of a project at school to get a bit experience using directX. I dont know why, but i couldnt stop the project and started programming at home aswell. At the moment i am trying to create some small AI. Of cause thats definatlly not easy, but thats my personal goal anyways. The topic could prolly fill multiple books hehe. I've got the walking part of my bots done so far. They walk along a scriped path. I am not working on the "aiming" of the bots. While programming that i hit on some math problem i couldnt solve yet. I hope of your input on this to help me get further. Concepts, ideas and everything else are highly appreciated. Problem: Calculate the position (D3DXVECTOR3) where the curve of the projectile (depends on gravity, speed), hit the curved of the enemys walking path (depends on speed). We assume that the enemy walks in a constant line. Known variables: float projectilSpeed = 2000 m/s //speed of the projectile per second float gravitation = 9.81 m/s^2 //of cause the gravity lol D3DXVECTOR3 targetPosition //position of the target stored in a vector (x,y,z) D3DXVECTOR3 projectilePosition //position of the projectile D3DXVECTOR3 targetSpeed //stores the change of the targets position in the last second Variabledefinition ProjectilePosition at time of collision = ProjectilePos_t TargetPosition at time of collision = TargetPos_t ProjectilePosition at time 0, now = ProjectilePos_0 TargetPosition at time 0, now = TargetPos_0 Time to impact = t Aim-angle = theta My try: Found a formular to calculate "drop" (Drop of the projectile based on the gravity) on Wikipedia: float drop = 0.5f * gravity * t * t The speed of the projectile has a horizontal and a vertical part.. Found a formular for that on wikipedia aswell: ProjectilVelocity.x = projectilSpeed * cos(theta) ProjectilVelocity.y = projectilSpeed * sin(theta) So i would assume this is true for the projectile curve: ProjectilePos_t.x = ProjectilePos_0.x + ProjectileSpeed * t ProjectilePos_t.y = ProjectilePos_0.y + ProjectileSpeed * t + 0.5f * gravity * t * t ProjectilePos_t.z = ProjectilePos_0.z + ProjectileSpeed * t The target walk with a constant speed, so we can determine his curve by this: TargetPos_t = TargetPos_0 + TargetSpeed * D3DXVECTOR3(t, t, t) Now i dont know how to continue. I have to solve it somehow to get a hold on the time to impact somehow. As a basic formular i could use: float time = distanz / projectileSpeed But that wouldnt be truly correct as it would assume a linear "Trajectory". We just find this behaivor when using a rocket. I hope i was able to explain the problem as much as possible. If there are questions left, feel free to ask me! Greets from germany, Frank

  Read the article

• Winforms: How to speed up Invalidate()?

  - by Pedery

  I'm developing a retained mode drawing application in GDI+. The application can draw simple shapes to a canvas and perform basic editing. The math that does this is optimized to the last byte and is not an issue. I'm drawing on a panel that is using the built-in Controlstyles.DoubleBuffer. Now, my problem arises if I run my app maximized on a big monitor (HD in my case). If I try to draw a line from one corner of the (big) canvas to the diagonally oposite other, it will start to lag and the CPU goes high up. Each graphical object in my app has a boundingbox. Thus, when I invalidate the boundingbox of a line that goes from one corner of the maximized app to the oposite diagonal one, that boundingbox is virtually as big as the canvas. When a user is drawing a line, this invalidation of the boundingbox thus happens on the mousemove event, and there is a clear lag visible. This lag also exists if the line is the only object on the canvas. I've tried to optimize this in many ways. If I draw a shorter line, the CPU and the lag goes down. If I remove the Invalidate() and keep all other code, the app is quick. If I use a Region (that only spans the figure) to invalidate instead of the boundingbox, it is just as slow. If I split the boundingbox into a range of smaller boxes that lie back to back, thus reducing the invalidation area, no visible performance gain can be seen. Thus I'm at a loss here. How can I speed up the invalidation? On a side note, both Paint.Net and Mspaint suffers from the same shortcommings. Word and PowerPoint however, seem to be able to paint a line as described above with no lag and no CPU load at all. Thus it's possible to achieve the desired results, the question is how?

  Read the article

• Daemon program that uses select() inside infinite loop uses significantly more CPU when ported from

  - by Jake

  I have a daemon app written in C and is currently running with no known issues on a Solaris 10 machine. I am in the process of porting it over to Linux. I have had to make minimal changes. During testing it passes all test cases. There are no issues with its functionality. However, when I view its CPU usage when 'idle' on my Solaris machine it is using around .03% CPU. On the Virtual Machine running Red Hat Enterprise Linux 4.8 that same process uses all available CPU (usually somewhere in the 90%+ range). My first thought was that something must be wrong with the event loop. The event loop is an infinite loop ( while(1) ) with a call to select(). The timeval is setup so that timeval.tv_sec = 0 and timeval.tv_usec = 1000. This seems reasonable enough for what the process is doing. As a test I bumped the timeval.tv_sec to 1. Even after doing that I saw the same issue. Is there something I am missing about how select works on Linux vs. Unix? Or does it work differently with and OS running on a Virtual Machine? Or maybe there is something else I am missing entirely? One more thing I am sure sure which version of vmware server is being used. It was just updated about a month ago though. Thanks.

  Read the article

• File transfer eating alot of CPU

  - by Dan C.

  I'm trying to transfer a file over a IHttpHandler, the code is pretty simple. However when i start a single transfer it uses about 20% of the CPU. If i were to scale this to 20 simultaneous transfers the CPU is very high. Is there a better way I can be doing this to keep the CPU lower? the client code just sends over chunks of the file 64KB at a time. public void ProcessRequest(HttpContext context) { if (context.Request.Params["secretKey"] == null) { } else { accessCode = context.Request.Params["secretKey"].ToString(); } if (accessCode == "test") { string fileName = context.Request.Params["fileName"].ToString(); byte[] buffer = Convert.FromBase64String(context.Request.Form["data"]); string fileGuid = context.Request.Params["smGuid"].ToString(); string user = context.Request.Params["user"].ToString(); SaveFile(fileName, buffer, user); } } public void SaveFile(string fileName, byte[] buffer, string user) { string DirPath = @"E:\Filestorage\" + user + @"\"; if (!Directory.Exists(DirPath)) { Directory.CreateDirectory(DirPath); } string FilePath = @"E:\Filestorage\" + user + @"\" + fileName; FileStream writer = new FileStream(FilePath, File.Exists(FilePath) ? FileMode.Append : FileMode.Create, FileAccess.Write, FileShare.ReadWrite); writer.Write(buffer, 0, buffer.Length); writer.Close(); }

  Read the article

• threading in Python taking up too much CPU

  - by KevinShaffer

  I wrote a chat program and have a GUI running using Tkinter, and to go and check when new messages have arrived, I create a new thread so Tkinter keeps doing its thing without locking up while the new thread goes and grabs what I need and updates the Tkinter window. This however becomes a huge CPU hog, and my guess is that it has to do somehow with the fact that the Thread is started and never really released when the function is done. Here's the relevant code (it's ugly and not optimized at the moment, but it gets the job done, and itself does not use too much processing power, as when I run it not threaded, it doesn't take up much CPU but it locks up Tkinter) Note: This is inside of a class, hence the extra tab. def interim(self): threading.Thread(target=self.readLog).start() self.after(5000,self.interim) def readLog(self): print 'reading' try: length = len(str(self.readNumber)) f = open('chatlog'+str(myport),'r') temp = f.readline().replace('\n','') while (temp[:length] != str(self.readNumber)) or temp[0] == '<': temp = f.readline().replace('\n','') while temp: if temp[0] != '<': self.updateChat(temp[length:]) self.readNumber +=1 else: self.updateChat(temp) temp = f.readline().replace('\n','') f.close() Is there a way to better manage the threading so I don't consume 100% of the CPU very quickly?

  Read the article

• determining the starting speed for an accelerated animation (in flash/actionscript but it's a math question)

  - by vulkanino

  This question burns my brain. I have an object on a plane, but for the sake of simplicity let's work just on a single dimension, thus the object has a starting position xs. I know the ending position xe. The object has to move from starting to ending position with an accelerated (acceleration=a) movement. I know the velocity the object has to have at the ending position (=ve). In my special case the ending speed is zero, but of course I need a general formula. The only unknown is the starting velocity vs. The objects starts with vs in xs and ends with ve in xe, moving along a space x with an acceleration a in a time t. Since I'm working with flash, space is expressed in pixels, time is expressed in frames (but you can reason in terms of seconds, it's easy to convert knowing the frames-per-second). In the animation loop (think onEnterFrame) I compute the new velocity and the new position with (a=0.4 for example): vx *= a (same for vy) x += vx (same for y) I want the entire animation to last, say, 2 seconds, which at 30 fps is 60 frames. Now you know that in 60 frames my object has to move from xs to xe with a constant deceleration so that the ending speed is 0. How do I compute the starting speed vs? Maybe there's a simpler way to do this in Flash, but I am now interested in the math/physics behind this.

  Read the article

• IPCop server slows down download speed

  - by noocyte

  I have an IPCop server running at home, been doing just fine for ~5 months, but last week I suddenly started getting time-outs and slow downloads from the 'net. I first thought that this was my ISP acting up, then I thought it might be one of my 3 switches or some of my cabling. In due order I've tested everything above and found them all to be working as they should. The only factor remaining is my IPCop server. Facts: I've got a 15/15 Mbit line (fiber) and I get ~15 Mbit upload, but only 0.5 Mbit download with the IPCop box as router (ISP router set in bridge mode). If I connect without the IPCop box (using the ISP router) I get ~12 Mbit upload and ~15 Mbit download. The load on the IPCop box appears to be light and it used to handle this traffic just fine 2 weeks ago. The memory usage is ~60%, I tried to restart it and test again, the memory fell to ~50% then (5 months of uptime). I'm thinking that one of my nics are busted, but I'm sort of perplexed that this could be the outcome; slow download but full speed upload. Anybody ever seen that happening before? Could it just be one of the nics that needs to be replaced? Will try that as soon as I can get my hands on a couple of new ones.

  Read the article

• cset as non-root to set cpu affinity for running processes

  - by RaveTheTadpole

  I've been playing with cset to set cpu affinity for running processes. I'm recreating the built-in "shield" function manually with set and proc, to add some subsets for specific threads of my application. I have a bash script that is calling cset to create the sets, and move the correct threads to the correct sets. It works when run with sudo. Now I'd like to make this script executable by another user, who does not have sudo powers. I trust this user enough to be responsible with cset, but don't want to open up the wide powers of root. I thought that CAP_SYS_NICE -- which is needed for sched_setaffinity, which I just assume cset must use -- on the script would be sufficient, but that didn't work. I tried extending CAP_SYS_NICE to the cset program (which is a thin python wrapper for the cset python library). No dice. The output of cap_to_text on my CAP_SYS_NICE'd scripts is "=cap_ipc_lock,cap_sys_nice,cap_sys_resource+eip" (it has ipc_lock and sys_resource for other reasons; I think only sys_nice is relevant). Any ideas?

  Read the article

• Still about SSD potentials...write and read speed

  - by Macroideal

  I have been working on SSD (solid state disk) for several months..Problems and Questions hit my head unexpectedly..Coz i am a virgin in ssd... Especially these days I was testing the write-read speed of ssd, which I was always caring.... however result turned out not good as I expected, or even worse Three kinds of read-write were implemented in my test read and write directly from and into ssd, with openning ssd as a whole device. in windows: _open("\\:g", ***).. It can be very tricky and hairy that you'd write a data with size of folds of 512, at the disk position of folds of 512bytes... So, If you wanto write just a byte or 4 bytes, you'v to write at least a whole sector one time. Read and write data from and into files located in SSD... Read and Write data from and into files in mechanical Disk I compared the pratices below...I found ssd sucks...the ssd performs worse than mechanical disk... so i am wondering where i can get the potential performance of ssd, since ssd is said to a substitute for mechanical disk in the future.. Nevertheless, I test ssd with a pro-hard-disk tools..ssd is like twice speedier than mechanical disk. So, why?

  Read the article

• Replacing stock Core 2 Duo heatsink fan (just the fan really) with a Dell CPU fan

  - by user647345

  My old heatsink fan broke and I'm trying to reconnect its plugs to a new fan. My Dell CPU fan has some custom Dell plug. I snipped the old fan's wire in half and kept the plug on the end of it. I want to connect it to the Dell fan wire to the plug. The motherboard is a P5Q-e, the stock Core 2 Duo fan was .20A and the dell is .70A. Is that going to matter? The wire from the fan has four wires, the wire with the plug has four wires. They share three, of the four colors: red, black, and blue. Dell's fourth wire is white, while the plug's fourth wire is yellow. Is it safe to assume that I just connect the yellow and the white plug together and match the rest up? I don't want to take any risk of damaging anything. It runs fine passively without a fan, but I have speedstep on, so I would like to use this fan and just fasten it to the heatsink with some twist ties and paperclips and call it a day.

  Read the article

• NFS high CPU usage

  - by user269836

  Hello, I have a very strange issue. I have next server: Intel(R) Xeon(TM) MP CPU 3.16GHz cat /proc/cpuinfo | grep proce | wc -l 8 free -m total used free shared buffers cached Mem: 28203 27606 596 0 10789 9714 -/+ buffers/cache: 7103 21100 Swap: 24695 0 24695 RAID card *-storage description: RAID bus controller product: MegaRAID vendor: LSI Logic / Symbios Logic physical id: 7 bus info: pci@0000:13:07.0 logical name: scsi2 version: 01 width: 32 bits clock: 66MHz capabilities: storage pm bus_master cap_list rom configuration: driver=megaraid latency=32 resources: irq:134 memory:d8ff0000-d8ffffff(prefetchable) memory:df600000-df60ffff(prefetchable) HDD: 10x148Gb SCSI U320 15k - RAID5 /dev/sdb1 807G 674G 93G 88% /storage /dev/sdb1 /storage ext4 defaults,usrjquota=aquota.user,grpjquota=aquota.group,jqfmt=vfsv0,noatime,nodiratime,noacl,errors=remount-ro 0 1 network cards ethtool -i eth0 driver: tg3 version: 3.116 firmware-version: 5704-v3.36, ASFIPMIc v2.36 bus-info: 0000:10:02.0 ethtool -i eth1 driver: tg3 version: 3.116 firmware-version: 5704-v3.36, ASFIPMIc v2.36 bus-info: 0000:10:02.0 ifconfig bond0 Link encap:Ethernet HWaddr 00:0f:1f:ff:d6:4d inet addr:192.168.15.71 Bcast:192.168.15.255 Mask:255.255.255.0 inet6 addr: fe80::20f:1fff:feff:d64d/64 Scope:Link UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1 RX packets:1062818202 errors:0 dropped:3918 overruns:0 frame:0 TX packets:1041317321 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:10000 RX bytes:258867684559 (241.0 GiB) TX bytes:396569192650 (369.3 GiB) this server running only nfs-kernel-server uname -a Linux nas2-backup 2.6.32-5-amd64 #1 SMP Sun Sep 23 10:07:46 UTC 2012 x86_64 GNU/Linux Debian 6. What do I have, once per day or two, LA goes up, it can be reached around LA: 40 but if I do: nfs-kernel-server restart. Every thing is OK. But on the next day or a little bit later, LA goes up again. Servers are connected to d-link dgs 1016d with 24 GBits ports. I have tried everything to find out what the problem is. Why it's happening, but still I can not resolve this issue. Any ideas on what is happening here?

  Read the article

• High load average due to high system cpu load (%sys)

  - by Nick

  We have server with high traffic website. Recently we moved from 2 x 4 core server (8 cores in /proc/cpuinfo), 32 GB RAM, running CentOS 5.x, to 2 x 4 core server (16 cores in /proc/cpuinfo), 32 GB RAM, running CentOS 6.3 Server running nginx as a proxy, mysql server and sphinx-search. Traffic is high, but mysql and sphinx-search databases are relatively small, and usually everything works blazing fast. Today server experienced load average of 100++. Looking at top and sar, we noticed that (%sys) is very high - 50 to 70%. Disk utilization was less 1%. We tried to reboot, but problem existed after the reboot. At any moment server had at least 3-4 GB free RAM. Only message shown by dmesg was "possible SYN flooding on port 80. Sending cookies.". Here is snippet of sar 11:00:01 CPU %user %nice %system %iowait %steal %idle 11:10:01 all 21.60 0.00 66.38 0.03 0.00 11.99 We know that this is traffic issue, but we do not know how to proceed future and where to check for solution. Is there a way we can find where exactly those "66.38%" are used. Any suggestions would be appreciated.

  Read the article

• Bash Parallelization of CPU-intensive processes

  - by ehsanul

  tee forwards its stdin to every single file specified, while pee does the same, but for pipes. These programs send every single line of their stdin to each and every file/pipe specified. However, I was looking for a way to "load balance" the stdin to different pipes, so one line is sent to the first pipe, another line to the second, etc. It would also be nice if the stdout of the pipes are collected into one stream as well. The use case is simple parallelization of CPU intensive processes that work on a line-by-line basis. I was doing a sed on a 14GB file, and it could have run much faster if I could use multiple sed processes. The command was like this: pv infile | sed 's/something//' > outfile To parallelize, the best would be if GNU parallel would support this functionality like so (made up the --demux-stdin option): pv infile | parallel -u -j4 --demux-stdin "sed 's/something//'" > outfile However, there's no option like this and parallel always uses its stdin as arguments for the command it invokes, like xargs. So I tried this, but it's hopelessly slow, and it's clear why: pv infile | parallel -u -j4 "echo {} | sed 's/something//'" > outfile I just wanted to know if there's any other way to do this (short of coding it up myself). If there was a "load-balancing" tee (let's call it lee), I could do this: pv infile | lee >(sed 's/something//' >> outfile) >(sed 's/something//' >> outfile) >(sed 's/something//' >> outfile) >(sed 's/something//' >> outfile) Not pretty, so I'd definitely prefer something like the made up parallel version, but this would work too.

  Read the article

• HP DL185 - very slow disk read speed

  - by fistameeny

  Hi, I have a HP DL185 G6 Server (12 disk model) with the following spec: Quad Core Xeon 2.27GHz 6GB RAM HP P212 RAID controller with battery backup 2 x 128GB 15K SAS 3.5" (RAID-1 for the operating system) 4 x 750GB 7.5K SAS 3.5" (RAID-5 for the data, 2TB usable space) The operating system is Ubuntu Server 9.10. Both drives have been formatted as EXT4. We are finding that read speed of the RAID-5 array is poor. Disk test results below: sudo hdparm -tT /dev/cciss/c0d1p1 /dev/cciss/c0d1p1: Timing cached reads: 15284 MB in 2.00 seconds = 7650.18 MB/sec Timing buffered disk reads: 74 MB in 3.02 seconds = 24.53 MB/sec For info, the RAID-1 array performs as follows: sudo hdparm -tT /dev/cciss/c0d0p1 /dev/cciss/c0d0p1: Timing cached reads: 15652 MB in 2.00 seconds = 7834.26 MB/sec Timing buffered disk reads: 492 MB in 3.01 seconds = 163.46 MB/sec We thought this was because with no battery, read/write cache is disabled. We have bought and installed the battery backup and have used the HP bootable CD to change the cache settings to 50% read / 50% write and check cache is enabled on the drives and the controller. Is there something I'm missing?

  Read the article

• Measure Upload Speed between a client and our server

  - by tresstylez

  We host a SAAS application specially customized for multiple clients. For one customer in particular -- they are reporting sporadic performance issues from various locations on their network, in particular UPLOADING documents through a form on our website. The client claims they have "bandwidth to spare" and that utilization of their "pipe" is so low that it MUST be our application, but our application has MANY clients and all features are working fine for all other clients. Interestingly enough -- DOWNLOADS (ie. just accessing the website, or downloading documents) is working fine. Speed test shows that they should get 1.2Mbps UP. So, a 3MB file should take 20 secs to upload. It takes 60+ seconds on their network. Sometimes even small files take OVER 10 minutes to upload or they timeout. Pings and Traceroutes don't show any abnormally long hops or response times. They claim other SAAS applications they use allow them to upload just fine. Both IT teams are working together to resolve this issue. What kind of data can I request from the clients to begin ruling things out. Seems like we need to somehow measure LATENCY of the networks involved or even at the switch level, we need to understand if packets are getting dropped somewhere and why. Where should I start? Any help is appreciated. I'll provide more info upon requests

  Read the article

• Bonnie does not provide speed for Sequential Input / Block

  - by Lqp1

  I'm using ProxmoxVE and I would like to run some benchmarks regarding performances of this product. One of these benchmarks is bonnie++ ; it runs very well in a VM (qemu-kvm) but when I run it in a conainer (openVZ), it does not provide me reading speed (only writing). I don't understand why... Does anyone know what's happenning ? VMs ans Containers are Debian 7.4. Here's the output of bonnie in the container: root@ct2:/# bonnie++ -u root Using uid:0, gid:0. Writing a byte at a time...done Writing intelligently...done Rewriting...done Reading a byte at a time...done Reading intelligently...done start 'em...done...done...done...done...done... Create files in sequential order...done. Stat files in sequential order...done. Delete files in sequential order...done. Create files in random order...done. Stat files in random order...done. Delete files in random order...done. Version 1.96 ------Sequential Output------ --Sequential Input- --Random- Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP ct2 1G 843 99 59116 8 60351 4 4966 99 +++++ +++ 2745 8 Latency 9558us 3582ms 527ms 1672us 936us 5248us Version 1.96 ------Sequential Create------ --------Random Create-------- ct2 -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete-- files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP 16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ Latency 19567us 358us 368us 107us 59us 25us 1.96,1.96,ct2,1,1401810323,1G,,843,99,59116,8,60351,4,4966,99,+++++,+++,2745,8,16,,,,,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,9558us,3582ms,527ms,1672us,936us,5248us,19567us,358us,368us,107us,59us,25us The filesystem for / is of type "simfs", which is a pseudo filesystem for openVZ. Maybe it's related to this issue but I can't find anyone with the same issue with bonnie and openVZ... Thanks for your help. Regards, Thomas.

  Read the article

• At what point does the performance gap between GPU & CPU become so great that the CPU is holding back a system?

  - by Matthew Galloway

  I know that generally speaking for gaming performance the GPU is the primary factor which holds back performance, with everything else such as RAM/motherboard/PSU/CPU being secondary in importance to the graphics card. But at some point the other components ARE going to be significant in holding back the whole system! For instance nobody would be silly enough to play modern games with 512MB RAM and the very latest graphics cards (such as an HD7970) as I bet the performance increase over such a system with only 512MB but a mid range card would be non-existent! Thus it would be a "waste" for such a person to buy any high end graphics card without resolving first the system's other problems. The same point applies to other components, such as if it only had a Pentium II a current high end graphics card would be wasted on it! So my core question is how do you determine at what point for your system is spending on extra GPU power be completely "wasted"? (also, a slightly more nuanced question is trying work out at what point might the extra graphics power not be "wasted" but would be "sub optimal" value for money, when the expenditure should then be split around graphics card and other components. As obviously a gamer shouldn't always just spend on upgrading the graphics card! But needs to balance it out)

  Read the article

< Previous Page | 49 50 51 52 53 54 55 56 57 58 59 60  | Next Page >