Search Results

Search found 415 results on 17 pages for 'bottleneck'.

Page 7/17 | < Previous Page | 3 4 5 6 7 8 9 10 11 12 13 14  | Next Page >

  • Questioning pythonic type checking

    - by Pace
    I've seen countless times the following approach suggested for "taking in a collection of objects and doing X if X is a Y and ignoring the object otherwise" def quackAllDucks(ducks): for duck in ducks: try: duck.quack("QUACK") except AttributeError: #Not a duck, can't quack, don't worry about it pass The alternative implementation below always gets flak for the performance hit caused by type checking def quackAllDucks(ducks): for duck in ducks: if hasattr(duck,"quack"): duck.quack("QUACK") However, it seems to me that in 99% of scenarios you would want to use the second solution because of the following: If the user gets the parameters wrong then they will not be treated like a duck and there will be no indication. A lot of time will be wasted debugging why there is no quacking going on until the user finally realizes his silly mistake. The second solution would throw a stack trace as soon the user tried to quack. If the user has any bugs in their quack() method which cause an AttributeError then those bugs will be silently swallowed. Once again time will be wasted digging for the bug when the second solution would simply give a stack trace showing the immediate issue. In fact, it seems to me that the only time you would ever want to use the first method is when: The block of code in question is in an extremely performance critical section of your application. Following the principal of "avoid premature optimization" you would only realize this of course, after you had implemented the safer approach and found it to be a bottleneck. There are many types of quacking objects out there and you are only interested in quacking objects that quack with a very specific set of arguments (this seems to be a very rare case to me). Given this, why is it that so many people prefer the first approach over the second approach? What is it that I am missing? Also, I realize there are other solutions (such as using abcs) but these are the two solutions I seem to see most often for the basic case.

    Read the article

  • SQLAuthority News – Presenting at Tech-Ed On Road – Ahmedabad – June 11, 2011 – Wait Types and Queues

    - by pinaldave
    I will be presenting in person on the subject SQL Server Wait Types and Queues at Ahmedabad on June 11, 2011. Here is the quick summary of the session. SQL Server Waits and Queues – Your Gateway to Perf. Troubleshooting Time: 11:15am – 12:15pm – June 11, 2011 Just like a horoscope, SQL Server Waits and Queues can reveal your past, explain your present and predict your future. SQL Server Performance Tuning uses the Waits and Queues as a proven method to identify the best opportunities to improve performance. A glance at Wait Types can tell where there is a bottleneck. Learn how to identify bottlenecks and potential resolutions in this fast paced, advanced performance tuning session. This session is based on my performance tuning Wait Types and Queues series. SQL SERVER – Summary of Month – Wait Type – Day 28 of 28 During the session there will be Quiz and those who gets right answer will get very interesting gifts from me. Do not miss a single minute of the event. We are also going to have two rock star speakers – Harish Vaidyanathan and Jacob Sebastian. Here is the details for the event: SQLAuthority News – Community Tech Days – TechEd on The Road – Ahmedabad – June 11, 2011 Reference: Pinal Dave (http://blog.SQLAuthority.com) Filed under: About Me, Pinal Dave, PostADay, SQL, SQL Authority, SQL Query, SQL Server, SQL Tips and Tricks, SQLAuthority Author Visit, T SQL, Technology

    Read the article

  • Shared Database Servers

    - by shivanshu.upadhyay
    As more enterprises consolidate their database environments to support private cloud initiatives, ISVs will have to deal with sceanrios where they need to run on a shared powerful database server like Exadata. Some ISVs are concerned about meeting SLAs for performance in a shared environment. Outside the virtualization world, there are capabilities of Oracle Database which can be used to prevent resource contention and guarantee SLA. These capabilities are - 1) Instance Caging - This guarantees the CPU allocated or limits the maximum number of CPUs (and so the number of Oracle processes) that an instance of Database can use simultaneously. With this feature, ISVs can be assured that their application is allocated adequate CPUs even if the database server is shared with other applications. 2) CPU Resource Allocation with Database Resource Manager - This allocates percentages of CPU time to different users and applications within a database. ISVs can use this feature to ensure that priority user or workloads within their application get CPU resources over other requirements. 3) Exadata I/O Resource Manager - The Database Resource Manager feature in Oracle Database 11g has been enhanced for use with Exadata. This allows the sharing of storage between databases without fear of one database monopolizing the I/O bandwidth and impacting the performance of the other databases sharing the storage. This can be used to ensure that I/O does not become a performance bottleneck due to poor design of other applications sharing the same server.

    Read the article

  • HD movies stutter.

    - by Absolute0
    I just put together a new system build in the hopes that all of my daily tasks would run smoothly and without any hiccups. Unfortunately I am still seeing some sound and sometimes video stuttering when playing HD movies in VLC (no problems with xvid/divx files). My setup is as follows: Intel core i5 750 quad core 2.66mhz 4GB ram asus p55 motherboard radeon hd 5570 video card 650gb 7200rpm western digital sata HDD 23" Nec ea23wmi monitor Operating System: Windows 7 What might be the main bottleneck that needs upgrading to fix my delays? Seems like the hard drive might be the problem but anything faster than 7200rpm is beyond my budget for a decent hard drive. Could it be anything else?

    Read the article

  • RAID P410i and P812 performance issues

    - by Alexey
    I'm having much trouble with I/O performance of HP DL360 server with two RAID controllers - P410i and P812, Windows Server 2008, 36 GiB RAM and 16 x Intel Xeon x5550. The server runs a bunch of tasks producing heavy sequential I/O, and after about 20-30 minutes of intensive work it looks like the tasks are stuck, not using CPU and with enough free memory (so this cannot be a bottleneck). The same tasks were running quite well on the older server (Windows Server 2003, 4 x Intel Xeon, 12 GiB RAM). RAID cache is present, write-cache battery is installed. Cache is configured as 25% readahead/75% writeback. The swap file resides on the logical disk served by P410i and other logical disks are on P812. Can someone tell me what can be the matter of this? Is this a hardware problem or misconfiguration?

    Read the article

  • What is this dm-0 device?

    - by Jeff Shattock
    While poking around trying to figure out why a Linux - Linux file transfer is running slower than I think it should, I stumbled across something I'm not familiar with. /dev/dm-0 seems to be my bottleneck, but I have no idea what it is. On my destination server, the iostat command shows a device at the bottom, /dev/dm-0, as being 100% utilized. This server has 6 disks in a mdadm raid5 set, with LVM running on top of it. Each of the underlying disks are sitting around 50% util. The transfer is writing to a logical volume located on this raidset. What is this /dev/dm-0 thing? Once I know what it is, maybe I can find how to increase its speed, or at least understand why its the speed that it is.

    Read the article

  • Diagnosing high CPU waiting

    - by Will
    I have a monitoring server that is running icinga/collectd/graphite with about 50 hosts. I have noticed high load/slugging performance on the box. If you take a look at top, you'll see: Cpu(s): 0.6%us, 0.2%sy, 0.0%ni, 7.6%id, 23.4%wa, 0.0%hi, 0.2%si, 0.0%st Notice the HUGE %wa value, which as far as I know means a network or disk bottleneck. ifconfig shows no dropping packets and there's not a ton of bandwidth going on, so that leaves disk issues, right? There's not a lot of disk writing going on either...iotop is reporting we're only writing a little over 1 MB per second and the RAID tool reports everything is A-OK and write caching is enabled. How do I go about trying to figure out how to fix this? UPDATE: iostat -x output is: avg-cpu: %user %nice %system %iowait %steal %idle 0.62 0.10 0.31 9.65 0.00 89.31 Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 0.21 33.34 83.55 16.54 1599.94 399.07 19.97 43.21 416.98 3.71 37.13

    Read the article

  • Internet connection slower than network connection speed

    - by Mike Pateras
    I've got a computer connected to a wireless router on a different floor. When I look at the network connection, I'm told the signal strength is low, and that I've got a connection of about 26mbps (often higher). However, my internet connection on that machine is very slow. Speedtests show it at about 1-2mbps, and it really shows when loading pages and video. I have fiber optic internet access, and the machine that's connected to the router/modem via cable gets the 20mbps on speed tests, and is extremely fast in every day use. My question is, is the advertised 26mbps+ connection speed perhaps inaccurate, and that my wireless bandwidth is the likely bottleneck here? Or is the signal strength what's key here? And what might I do about this? Power cycling the router helped a bit, a speed test went as high as 6mbps after doing that.

    Read the article

  • FreeBSD: Samba performance over GBit-Ethernet

    - by Axel Gneiting
    I'm using a FreeBSD NAS with RAID-Z. I can read ~300MB/s from the ZFS disks to /dev/null on the box, but only get about 50MB/s over GBit-Ethernet with SMB to Windows 7 (Samba 3.5.6). Both systems have Intel-PCIe-NICs and are connected directly. Samba is configured to use AIO and I already tried to tune TCP/IP: kern.ipc.maxsockbuf=16777216 net.inet.tcp.sendspace=1048576 net.inet.tcp.recvspace=1048576 net.inet.tcp.sendbuf_max=8388608 net.inet.tcp.recvbuf_max=8388608 net.inet.tcp.delayed_ack=0 Any ideas what's causing the bottleneck? I think the link should handle 100 MB/s easily.

    Read the article

  • Stackify Gives Devs a Crack at the Production Server

    - by Matt Watson
    Originally published on SDTimes.com on 7/9/2012 by David Rubinstein.It was one of those interviews where you get finished talking about a company’s product, and you wonder aloud, “Well, THAT makes sense! Why hasn’t anyone thought of that before?” Matt Watson, CEO of Kansas City, Mo.-based startup Stackify, was telling me that the 10-person company is getting ready to launch its product in August (it’s in beta now) that will give developers an app-centric look into production servers so they can support and troubleshoot apps and fix bugs. Of course, this hasn’t happened in the past because of the security concerns of IT administrators, and a decided lack of expertise on the part of developers. Stackify installs on a server and acts like a proxy for developers, collecting data about the environment, discovering all the applications, scanning for config file changes, and doing server monitoring. “We become the central point that developers can see everything they need to know about their applications,” he said. “Developers can look at the files that are deployed, and query databases in a safe way.”  In his words:“The big thing we’re hoping is just giving them (developers) visibility. Most companies want to hire the junior developers that they pay $50,000 a year right out of college to do application support and troubleshooting and fix bugs, but those people don’t have access to production servers to troubleshoot. It becomes very difficult for them to do their job, so they end up spending all of their day bugging the senior developers, the managers or the system administrators to track down this stuff, which creates a huge bottleneck. And so what we can do is give that visibility to those lower-level people so that they can do this work and free up the higher-level people so they can be working on the next big thing.”Stackify itself might just prove to be the next big thing.

    Read the article

  • monitoring TCP/IP performance on Solaris

    - by Andy Faibishenko
    I am trying to tune a high message traffic system running on Solaris. The architecture is a large number (600) of clients which connect via TCP to a big Solaris server and then send/receive relatively small messages (.5 to 1K payload) at high rates. The goal is to minimize the latency of each message processed. I suspect that the TCP stack of the server is getting overwhelmed by all the traffic. What are some commands/metrics that I can use to confirm this, and in case this is true, what is the best way to alleviate this bottleneck?

    Read the article

  • What's the best external hard drive configuration for a software developer laptop?

    - by Dan
    I've got a Dell laptop that I use as a software developer box at work and find that the drive is usually the bottleneck. I'd like to hook up two 10k RPM drives that are striped for performance. I've looked for drive cases with RAID but there don't seem to be very many choices and I'm worried about compatibility with the drives (preferably SATA 2). Also I don't have a SATA connection on my machine so it'll have to USB 2.0 for now. Am I headed down the right path or am I missing a much simpler configuration?

    Read the article

  • How to design highly scalable web services in Java?

    - by Kshitiz Sharma
    I am creating some Web Services that would have 2000 concurrent users. The services are offered for free and are hence expected to get a large user base. In the future it may be required to scale up to 50,000 users. There are already a few other questions that address the issue like - Building highly scalable web services However my requirements differ from the question above. For example - My application does not have a user interface, so images, CSS, javascript are not an issue. It is in Java so suggestions like using HipHop to translate PHP to native code are useless. Hence I decided to ask my question separately. This is my project setup - Rest based Web services using Apache CXF Hibernate 3.0 (With relevant optimizations like lazy loading and custom HQL for tune up) Tomcat 6.0 MySql 5.5 My questions are - Are there alternatives to Mysql that offer better performance for what I'm trying to do? What are some general things to abide by in order to scale a Java based web application? I am thinking of putting my Application in two tomcat instances with httpd redirecting the request to appropriate tomcat on basis of load. Is this the right approach? Separate tomcat instances can help but then database becomes the bottleneck since both applications access the same database? I am a programmer not a Db Admin, how difficult would it be to cluster a Mysql database (or, to cluster whatever database offered as an alternative to 1)? How effective are caching solutions like EHCache? Any other general best practices? Some clarifications - Could you partition the data? Yes we could but we're trying to avoid it. We need to run a lot of data mining algorithms and the design would evolve over time so we can't be sure what lines of partition should be there.

    Read the article

  • Component-wise GLSL vector branching

    - by Gustavo Maciel
    I'm aware that it usually is a BAD idea to operate separately on GLSL vec's components separately. For example: //use instrinsic functions, they do the calculation on 4 components at a time. float dot = v1.x*v2.x + v1.y * v2.y + v1.z * v2.z; //NEVER float dot = dot(v1, v2); //YES //Multiply one by one is not good too, since the ALU can do the 4 components at a time too. vec3 mul = vec3(v1.x * v2.x, v1.y * v2.y, v1.z * v2.z); //NEVER vec3 mul = v1 * v2; I've been struggling thinking, are there equivalent operations for branching? For example: vec4 Overlay(vec4 v1, vec4 v2, vec4 opacity) { bvec4 less = lessThan(v1, vec4(0.5)); vec4 blend; for(int i = 0; i < 4; ++i) { if(less[i]) blend[i] = 2.0 * v1[i]*v2[i]; else blend[i] = 1.0 - 2.0 * (1.0 - v1[i])*(1.0 - v2[i]); } return v1 + (blend-v1)*opacity; } This is a Overlay operator that works component wise. I'm not sure if this is the best way to do it, since I'm afraid these for and if can be a bottleneck later. Tl;dr, Can I branch component wise? If yes, how can I optimize that Overlay function with it?

    Read the article

  • Caching DNS server (bind9.2) CPU usage is so so so high.

    - by Gk
    Hi, I have a caching-only dns server which get ~3k queries per second. Here is specs: Xeon dual-core 2,8GHz 4GB of RAM Centos 5x (kernel 2.6.18-164.15.1.el5PAE) bind 9.4.2 rndc status: recursive clients: 666/4900/5000 About 300 new queries (not in cache) per second. Bind always uses 100% on one core on single-thread config. After I recompiled it to multi-thread, it uses nearly 200% on two core :( No iowait, only sys and user. I searched around but didn't see any info about how bind use CPU. Why does it become bottleneck? One more thing, here is RAM usage: cat /proc/meminfo MemTotal: 4147876 kB MemFree: 1863972 kB Buffers: 143632 kB Cached: 372792 kB SwapCached: 0 kB Active: 1916804 kB Inactive: 276056 kB I've set max-cache-size to 0 to make sure bind can use as much RAM as it want, but it always stop at ~2GB. Since every second we got not cached queries so theoretically RAM must be exhausted but it wasn't. Do you have any idea? TIA, -Gk

    Read the article

  • Recommended Setup

    - by Chris Ryan
    I have been running into issue with my MSSQL Database setup with speed. Here is my scenario. About 100M Rows Average: 1k Updates Per Second Hard Drives: RAID 10 SSD MDF --Active Time: 0 Log Drives: 1 SSD LDF - Simple Recovery --Active Time 99.9 --Queue: 8 I do not need a back up of the log so it is set to simple recovery but my bottleneck is still at my log. I get high WAITLOG times and thus it can not update any faster. I can't do bulk updates/transactions and each update needs to be one at a time. Is my only option to increase write performance of the log drives, add a RAID drives? Any suggestions on increasing the performance?

    Read the article

  • Harvard vs. Von Neumann architecture

    - by user32569
    Hi. Our teacher told us, that Harvard architecture is the most evolved and produced architecture today and towards future. but I thing becouse os massive averhead of x86 and Von Neumann nased ARM systems that actually Von Neumann is the most used architecture today. Yes, MCUs with Harvard are even more produced, but since they all have just minor purpose (compared to x86 and ARM based) that Von Neumann is actually the one. Or is it really Harvard? And second, I know this is strange question, but does any architecture combining both exists? to have separate memory for data and programs, therefore faster instruction processing, but still able to work with these as Von Neumann? To be able o load amd unload programs to program memory on the fly? Isnt this the way the x86 should have go? Or would there be some bottleneck that pure Von neumann solves? Thanks.

    Read the article

  • How to speed up apache

    - by Zen_silence
    We have a server with 8Cores, 16GB of RAM and RAID 0 SAS 10K drives. Our goal is to use this to serve a fairly simple php application quickly. We have tested all other components and we think we have narrowed it down to apache is our bottleneck. I am no apache guru I have done some research and tested a couple things but when i test with JMeter launching 100 concurrent connections against the server the first 10 - 20 come back quickly 30 - 100ms but the rest take between 1000ms to 3000ms. Anyone have any ideas on what to change in our apache config to make this faster right now its a vanilla install of apache.

    Read the article

  • How to build a "traffic AI"?

    - by Lunikon
    A project I am working on right now features a lot of "traffic" in the sense of cars moving along roads, aircraft moving aroun an apron etc. As of now the available paths are precalculated, so nodes are generated automatically for crossings which themselves are interconnected by edges. When a character/agent spawns into the world it starts at some node and finds a path to a target node by means of a simply A* algorithm. The agent follows the path and ultimately reaches its destination. No problem so far. Now I need to enable the agents to avoid collisions and to handle complex traffic situations. Since I'm new to the field of AI I looked up several papers/articles on steering behavior but found them to be too low-level. My problem consists less of the actual collision avoidance (which is rather simple in this case because the agents follow strictly defined paths) but of situations like one agent leaving a dead-end while another one wants to enter exactly the same one. Or two agents meeting at a bottleneck which only allows one agent to pass at a time but both need to pass it (according to the optimal route found before) and they need to find a way to let the other one pass first. So basically the main aspect of the problem would be predicting traffic movement to avoid dead-locks. Difficult to describe, but I guess you get what I mean. Do you have any recommendations for me on where to start looking? Any papers, sample projects or similar things that could get me started? I appreciate your help!

    Read the article

  • How to cache streaming video and silverlight with squid windows reverse proxy

    - by V. Romanov
    We have an intranet web server running a silverlight application (ACTUS media monitor if anyone cares to know). The server is used to record video and stream it to clients through a CDN solution. We want to put a reverse proxy in between the server and CDN provider in order to remove the office network bottleneck that's currently strangling us. I've set up SQUID for windows on a separate machine outside the network using squid BasicAccelerator configuration setting. It seems to work as far as the reverse proxy is concerned, requests are forwarded and the application is working but it doesn't seem to cache anything (no space is used on the drive where squid is installed). I found to explicit setting to turn caching on in squid, so i assume it's on by default. Perhaps I need some other trick to make the video and/or silverlight cacheable? Any help will be appreciated. Any info you need to help me will be provided at once. Thanks in advance!

    Read the article

  • replacement drive cage for power edge R710

    - by bumble_bee_tuna
    Hi I'm performance tuning a DB server its a Dell R710 there is a very significant I/O bottleneck. Unfortunately the server was purchased with the 6 x 3.5 inch sata configuration which doesn't give me the leeway I need to address the issue. Before going to DAS does anyone know if it is possible to purchase a replacement front drive enclosure ? I know the server is configurable with something like 12 or 16 2.5 inch drives and it appears to be modular ? I tired contacting dell but the offshore parts department reps are not very bright lol. Thanks.

    Read the article

  • Is there a RAR extractor (for multiple rar files like .r00 etc.) that will use all of my quad cores?

    - by Christopher Done
    I've got a quad core Intel processor. I've got a big file split into little ones as RAR files, foo.r00, foo.r01, etc. which the RAR program extracts into one file/directory. Is there a RAR program that I can specify like "use four cores" in the extract process? At the moment it sits there using 100% of one core. I recognise the bottleneck might be my hard drive anyway, but I don't see a lot of HD usage and suspect the decompression process is more intensive than waiting on I/O. For example, GNU Make accepts a (-j, I think) argument to tell it how many cores to use, which I used to compile PHP 6 really quickly.

    Read the article

  • 32 vs 64-bit software for same machine?

    - by GorillaSandwich
    What is the difference between 32 and 64-bit software? My understanding is that 64-bit can use more RAM, if it's available, because it has a larger address space for it. Is this correct? And, specifically: If I have a 64-bit operating system with lots of RAM, and I install, say, the 32-bit version of MySQL instead of the 64-bit version, will it be unable to use all the available RAM and therefore run slower than the 64-bit version might on the same machine (assuming RAM becomes the bottleneck before processing speed or disk access speed or whatever)? If I have a 32-bit operating system and I install a 64-bit piece of software on it, will it (probably) fail to run?

    Read the article

  • Solaris TCP/IP performance tuning

    - by Andy Faibishenko
    I am trying to tune a high message traffic system running on Solaris. The architecture is a large number (600) of clients which connect via TCP to a big Solaris server and then send/receive relatively small messages (.5 to 1K payload) at high rates. The goal is to minimize the latency of each message processed. I suspect that the TCP stack of the server is getting overwhelmed by all the traffic. What are some commands/metrics that I can use to confirm this, and in case this is true, what is the best way to alleviate this bottleneck? PS I posted this on StackOverflow originally. One person suggested snoop and dtrace. dtrace seems pretty general - are there any additional pointers on how to use it to diagnose TCP issues?

    Read the article

  • Limiting calls to WCF services from BizTalk

    - by IntegrationOverload
    ** WORK IN PROGRESS ** This is just a placeholder for the full article that is in progress. The problem My BTS solution was receiving thousands of messages at once. After processing by BTS I needed to send them on via one of several WCF services depending on the message content. The problem is that due to the asynchronous nature of BizTalk the WCF services were getting hammered and could not cope with the load. Note: It is possible to limit the SOAP calls in the BtsNtSvc.exe.Config file but that does not have the desired results for Net-TCP WCF services. The solution So I created a new MessageType for the messages in question and posted them to the BTS messaeg box. This schema included the URL they were being sent to as a promoted property. I then subscribed to the message type from a new orchestraton (that does just the WCF send) using the URL as a correlation ID. This created a singleton orchestraton that was instantiated when the first message hit the message box. It then waits for further messages with the same correlation ID and type and processs them one at a time using a loop shape with a timer (A pretty standard pattern for processing related messages) Image to go here This limits the number of calls to the individual WCF services to 1. Which is a good start but the service can handle more than that and I didn't want to create a bottleneck. So I then constructed the Correlation ID using the URL concatinated with a random number between 1 and 10. This makes 10 possible correlation IDs per URL and so 10 instances of the singleton Orchestration per WCF service. Just what I needed and the upper random number is a configuration value in SSO so I can change the maximum connections without touching the code.

    Read the article

< Previous Page | 3 4 5 6 7 8 9 10 11 12 13 14  | Next Page >