I am running a large scale ERP system on the following server configuration. The application is developed using AngularJS and ASP.NET 4.5
Dell PowerEdge R730 (Quad Core 2.7 Ghz, 32 GB RAM, 5 x 500 GB Hard disk, RAID5 configured) Software: Host OS is VMWare ESXi 6.0 Two VMs run on VMWare ESXi .. one is Windows Server 2012 R2 with 16 GB memory allocated ... this contains IIS 8 server with my application code Another VM is also Windows Server 2012 R2 with SQL Server 2012 and 16 GB memory allocated .... this just contains my application database.
You see, I separated the application server and database server for load balancing purposes.
My application contains a registration module where the load is expected to be very very high (around 10,000 visitors over 10 minutes)
To support this volume of requests, I have done the following in my IIS server -> increase request queue in application pool length to 5000 -> enable output caching for aspx files -> enable static and dynamic compression in IIS server -> set virtual memory limit and private memory limit of each application pool to 0 -> Increase maximum worker process of each application pool to 6
I then used gatling to run load testing on my application. I injected 500 users at once into my registration module.
However, I see that only 40% / 45% of my RAM is being used. Each worker process is using only a maximum amount of 130 MB or so.
And gatling is reporting that around 20% of my requests are getting 403 error, and more than 60% of all HTTP requests have a response time greater than 20 seconds.
A single user makes 380 HTTP requests over a span of around 3 minutes. The total data transfer of a single user is 1.5 MB. I have simulated 500 users like this.
Is there anything missing in my server tuning? I have already tuned my application code to minimize memory leaks, increase timeouts, and so on.
There is a known issue with the newest generation of PowerEdge servers that use the Broadcom Network Chip set. Apparently, the "VM" feature for the network is broken which results in horrible network latency on VMs.
Head to Dell and get the most recent firmware and Windows drivers for the Broadcom.
Head to VMWare Downloads and get the latest Broadcom Driver
As for the worker process settings, for maximum performance, you should consider running the same number of worker processes as there are NUMA nodes, so that there is 1:1 affinity between the worker processes and NUMA nodes. This can be done by setting "Maximum Worker Processes" AppPool setting to 0. In this setting, IIS determines how many NUMA nodes are available on the hardware and starts the same number of worker processes.
I guess the 1 caveat to the answer you received would be if your server isn't NUMA aware/uses symmetric processing, you won't see those IIS options under CPU, but the above poster seems to know a good bit more than I do about the machine. Sorry I don't have enough street cred to add this as a comment. As far as IIS you may also want to make sure your app pool doesn't use default recycle conditions and pick a time like midnight for recycle. If you have root level settings applied the default app pool recycling at 29 hours may also trigger garbage collection against your child pool/causing delays even in concurrent gc where it sounds like you may benefit a bit from Gcserver=true. Pretty tough to assess that though.
Has your sql server been optimized for that type of workload? If your data isn't paramount you could squeeze faster execution times with delayed durability, then assess queries that are returning too much info for async io wait types. In general there's not enough here to really assess for sql optimizations, but if not configured right (size/growth options) you could be hitting a lot of timeouts due to growth, vlf fragmentation, etc.
Related
I have a Windows Server 2016 machine that runs a server program, there're about 2.2k concurrent requests per second. The server program only costs the server 25% cpu and 25% memory and 30% bandwidth. It's written in c++, just like the boost example. it just does some calculation and return the result to client in TCP, and it doesn't use the disk.
But it's very lag, I can see the lag not only from my clients, but also from the Remote Desktop Connection, it takes about 10 seconds to establish an RDP connection, and it's very quick(less than 2 seconds) if I close the server program.
I guess some resources on my server is exhausted. But how can I find it, is there any tool can profile the system to find the bottleneck?
Update
The server program uses all cores averagely by running 8 threads on 8 cores, I did take care about this, it's confirmed in Task Manager, all 8 cores used nearly the same.
I found the problem is: I'm using a sqlite3 database(my.db) to log all the client access, the server becomes more lag when the .db grows. Now it is 1.2Gb, which causes the lag.
Then I tried:
Keep the 1.2Gb .db, just load it once at startup to read some configuration, stop recording new log, no read/write access while server is running, but it's still lag.
Execute delete from log_table and vacuum to delete the previous log and reduce the .db size to 16k. Then lag problem is gone, client request becomes very quick.
Question
Why the large database can cause the whole server lag? Not only for the server itself, but also affect other app like RDP connection, even the load is low?
Server Environment
Windows Server 2016
cpu: 8 cores (25% used)
memory: 16Gb (25% used)
disk: 40Gb (30% used)
server program written in c++ with boost coroutine
sqlite3 database with PRAGMA journal_mode=WAL; enabled.
Install the sysinternals tools.
Launch procexp.exe (Process Explorer ) - use process explorer to find out memory and disk usage for your process, and other.
Use resmon ( Win+R then type "resmon" ) to monitor the network bandwidth when your program is running and when it's not.
I have an interesting problem with how Windows and .Net manage memory for Asp.Net applications that I can't explain myself. The problem is that I have a big Asp.Net application that after starts up can take about 1 GB memory according Resource Manager. We tried to test how many instances of the application we can run at the same time on a single machine with 14-16 GB memory.
First test is with an Azure Windows 2016 server with 8 vCPUs, 14 GB RAM, HDD.
After a few instances:
After 30 instances:
As you can see, private byes and working set of some instances reduced a lot. Based on what I read from how memory is managed (aka working set, physical memory, virtual memory, page files...), I can understand how the OS can take physical memory away from some idle processes for the others that are in need. So far so good.
Then we tested the same scenario with another Azure Windows 2016 server with 4 vCPUs, 16 GB RAM, but this one uses SSD.
After about 20 instances, we got OutOfMemoryException:
The key difference I could see is that memory of all those w3wp processes were still high. In other words, they were not reduced as in the test above.
My question is why the behaviors were different? What prevented the second cases from saving memory to page file (my guess!) and thus caused OutOfMemoryException?
Checking pagefile setting showed us that it was stilled enabled in "System managed size" mode but somehow Windows refused to use it for the w3wp processes. We tried to change it to custom size and set it to 20 GB and everything started working again as expected. I must admit that I still don't know why Windows 2016 behaves like that when SSD is used though.
I have a Windows Server 2012 R2 with 128Gb of memory and INtel Xeon E5 2650 v3.
Any time the memory usage for the App Pool goes close to 2.4Gb (according to Task Manager) I'm starting to get Out of memory exceptions and I have to recycle the app pool.
The Virtual Memory Limit is set to 0 for this app pool.
The overall server memory usage is always under 25%.
How do I increase the app pool memory limit?
How do I make it so that is not running as 32-bit application? Is this happening because I enabled Enable 32-bit applications? I actually need this because I have a managed assembly which is 32-bit only.
What is other solution for this? Don't say check your code, I need for certain cases that memory usage amount.
Thx
32 bit applications can address a maximum of 2 GB. Also see https://msdn.microsoft.com/en-us/library/windows/desktop/aa366778(v=vs.85).aspx
We're working towards setting up a web farm and storing sessions anywhere else but on the local state server is a required step. So I have tried 2 different options:
Storing sessions on a redis server using Microsoft.Web.RedisSessionStateProvider
Storing sessions on a SQL server 2014 standard edition
Both servers were dedicated to the task, in the same network than the web server and good enough spec. Each attempt only raised the CPU slightly on these servers which seems to indicate the bottleneck isn't there.
But for each attempt, changing the web.config from using the local state server to using redis or MSSQL caused the web server to be extremely sluggish and made our web site unbearably slow at times.
Some potentially useful facts:
When using the local state server, the web server CPU averages 50% CPU
Size of a session could be anything between 200 and 1500 Kb per user (we store a few reference tables in order to do less queries on the database)
Our session state server currently uses 500 Mb memory at peak time
We serve about 3,000 sessions daily, for an average time of 13 minutes per session
Web server is a 4 vCPU 2.4GHz 16GB ram, running IIS, ASP.net 4.0
Everything is hosted on AWS
Very happy to provide more metrics if required. Any idea why could cause it?
I am running IIS6 on Windows 2003 Server 32 bit. I have read that IIS6 has a maximum virtual memory limit of 2gb (3gb with the 3gb switch fipped).
What I am unclear on is whether this means all ASP.NET sessions have 2gb between them or 2gb each.
So if I have a session variable storing 200kb and have 10,000 active sessions am I going to be hitting up against this 2gb limit?
In general the advice is to leave these options unticked for ASP.NET applications, it affects how quickly the appPool recycles more information here summary below:
Physical and Virtual memory: This section is for recycling application pools which consume too much memory. Focusing on physical I typically like to limit app pools around 800MB to 1200 MB max on a 32 bit app with very few app pools depending on the number and amount of memory. On a server with 2 GB RAM I'd set it at around 800MB max. On a 4GB of RAM server around 1GB and more if more with a max around 1200. On a 64 bit web front end with 8-16 GB memory I've heard of settings of 2GB of RAM or even allowing it to let it ride, rather than limiting it.
You really need to profile it since these can really grow to process and cache. The greater the amount of memory and the greater the load the higher the worker process will grow. When people ask about configuring the app pool, this is where they are usually asking what the numbers should be. What you are doing here is explicitly limiting the app pool from consuming more memory.
Notice this setting is on the recycle tab, there's a reason for that. When an app pool reaches the max it isn't like the max processor setting. It will cycle the worker process which is like a tiny reboot or similar to an iisreset, but not since sometimes we want this to happen so we can release our memory. You really don't want to cycle more than a couple of times per 24 hour period in an ideal world. I've heard of some trying to cycle right before the morning peak occurs so they have the most amount of memory available, then a cycle right at the end of the day before the backups or crawling begins.
Basically the recommendation is not setting a limit (leave the options unchecked) because once the limit is hit IIS will recycle the application pool causing all active users to be temporarily disconnected from the site. You users will likely receive an HTTP 500 while the application pool recycles and then once it's back there will be a delay while the application pool warms up.
Sessions
For an application of any size do not use InProc (stored in memory) sessions use state server or SQL server to store your sessions. http://msdn.microsoft.com/en-us/library/ms178586.aspx
Conclusion
It really depends on the profile of your application, if your expecting 10,000 active sessions though don't use InProc, don't use IIS6 and don't use a 32 bit server.