DCSIMG

 Subscribe in a reader

Performance - Guy kolbis

Browse by Tags

All Tags » Performance (RSS)
I thought you might find that interesting. According to STP magazine testers choose VSTS and gave it the top spot in two categories: Dot NET Performance Testing. Integrated Test/Performance Suite. This is cool! One quick note is that in the Dot Net Performance Testing category, VSTS won over load runner! More Info Visual Studio Team Test Edition wins STP Magazine's Tester’s Choice poll …
נשלח על ידי kolbis | 3 תגובה/ות
Problem Space Network utilization is an important performance indicator, in fact, there are four critical performance indicators: CPU, Memory, IO and Network. Having said that, how do you know that is your network utilization? I am sure you know that the Task Manager allow you to see the network utilization In Vista you can also use the Reliability and Performance Monitor But, what about windows performance counters? How can it be that Microsoft did not publish any performance counter for Network...
נשלח על ידי kolbis | 1 תגובה/ות
Ed Glass from the MS Load Test Team wrote a post: Make Your Performance Work Count: The 20% Rule Basically, the idea behind the 20% rule is that the end users are "blind" to changes that within the limit of 20%. If you have a regression issue that is lower than 20% you might want to think if business wise you want to improve it. Here is a great sentence from his post: "If it is fast and ugly, they will use it and curse you; if it is slow, they will not use it." (David Cheriton...
נשלח על ידי kolbis
In one of my previous posts, I have talked about " Reaching Capacity in ASP.NET - Tips ". One thing I forgot to talk about is : what is capacity? As always, I wanted first to give you the "dictionary" translation...so I "googled" (or "lived") it and came up with this: " Capacity testing is complementary to load testing and it determines your server's ultimate failure point, whereas load testing monitors results at various levels of load and traffic...
נשלח על ידי kolbis | 6 תגובה/ות
There are several ways to cache HTTP responses and ETag is one of them. ETag is a nice feature that can sometimes help boosting the performance for a web application. Basically the idea behind ETag is to prevent redownloading and processing a page that has not been modified. The origin server specifies the ETag using the response header: HTTP/1.1 200 OK Last-Modified: Tue, 12 Dec 2006 03:03:59 GMT ETag: "10c24bc-4ab-457e1c1f" Content-Length: 12195 The browser will check for the expiration...
נשלח על ידי kolbis | 2 תגובה/ות
תגים:,
Here is a question I here often: "We want to load test our application, what do we need?" My answer is "very simple": "What are the goals for our load test?" That is when I usually here the sentence: "I want to see how my application operates under load..." Well, this is not a very good answer, or at least not satisfying one. "Why?" When we load test the application we can check several aspects that require deferent preparations. Usually we would...
נשלח על ידי kolbis
תגים:
When running a load test and collecting data, the Avg. Response Time is automatically collected for you. It is the average for all the responses occurred during the load test execution. However, sometimes you need the actual response time for each test/transaction. In order to get this information you must configure the load test to collect Timing information. When you select this, the actual values will be stored in the load test database. However, as for now (vs2008) there is no display for the...
נשלח על ידי kolbis | 2 תגובה/ות
Capacity testing is basically trying to break our application. So, how do we break it? Take the load formula and multiply it by a factor. For example if the load includes invoking 100 requests/sec, try interactively to multiply it by a constant (1,2,3…) factor starting with 2 and continue while you did not reach your capacity. Run the test and wait for failed requests. § TIP: Set the test for run duration of at least 30 minutes. If a request fails, check what the critical resource that caused it...
נשלח על ידי kolbis | 8 תגובה/ות
I have encountered a strange bug while using Visual Studio Team System 2008 Load Test. I configured a load test and executed it. Here is the test run result: As you can see, I have selected to monitor the Bytes Total/Sec performance counter during the load test execution. The strange thing about it is the following result:   As you can see,the Min value is 1,826,176 bytes total/sec and the max value is 3,506,534 bytes total/sec, however the average is 81,056. This must be wrong!!!
נשלח על ידי kolbis | 2 תגובה/ות
I know that those of you who visited the DevAcademy2 had a problem selecting the lectures to attend. So, I have been asked to repeat my presentation on: "Application Performance Analysis With VSTS" However, this time I have much more time.... :) So, I am going to take my presentation and expend it further. So, what is on our menu? PDD (Performance Driven Development). Profiling our application with Sampling and Instrumentation. Investigate memory allocation and objects lifetime. Collecting...
נשלח על ידי kolbis
In some cases you would like to monitor how your remote machines perform. Lets say that you developed a web application that was deployed on two machines. Machine A hosts the web application in the IIS machine B will host the SQL Services. Either one of the machine holds vital information regarding the performances for you application, thus it is important to monitor both. So, how do you monitor remote computers using the performance monitor? Basically you first need to allow remote access to the...
נשלח על ידי kolbis
תגים:
How can you tell which method consumes most of your CPU? I had an application that consumed a lot of CPU. I have noticed it through the performance monitor. The performance monitor does not give you an indication on what is the most consuming method, rather it give you a general idea on how you critical resources behave during the application run. However you can use the Visual Studio Profiler or other tools, but believe me, this will take more than one minute. Actually it can take you a few hours...
נשלח על ידי kolbis
תגים:
Here is a strange scenario I encountered. One of my customers complained that the number of threads is continuously increasing. So, in order to check it, I opened the Pefmon and selected the System category that includes both processes count and the threads count. I monitored it for a while and could not detect any increase in the total threads count. Then I opened the Objects category in the Perfmon and selected the Processes and the Threads objects. At first glance I could see that the System counters...
נשלח על ידי kolbis
תגים:
In the following cast, I will be talking about the transactions for both unit tests and web tests. Transactions specifies one or more steps that are grouped together, in the web test context those steps are a bunch of URLs, where in unit test it is a bunch of code that is marked as a group. So, why do we need it? I will try and explain it using an example. Lets say that you have a really complicated scenario that must be executed within a unit test. The scenario requires a long initialization and...
"What is pages/sec performance counter ?" The pages/sec performance counter is a memory related counter that measures the hard page faults. A hard page fault occurs when a memory page is not in the immediate memory and needs to be fetched from the disk. It is a very important counter that can potentially help us find bottlenecks. From my experience, a good warning threshold value for the pages/sec counter would be 10 and critical threshold value would be 20. "What is page faults/sec performance counter...
נשלח על ידי kolbis
תגים:
More Posts Next page »