i'll try to compare the same script done in Http/html with TruClient. In both of the scenarios, it has same think time/wait time, same number of vusers, same pacing.
Is it possible that they have approximately same time for each transactions but they are so different in term of total number of passed transactions?
Ty in advance
In web HTTP/HTMl protocol, Response time = Processing time + Latency (time taken by network while transferring data).
In Truclient protocol, Response time = Processing time + Latency + Rendering time
Hence you will found a difference between both response times.
And execution times will differ in both protocols, hence total number of passed transactions also vary.
The question comes on what are you trying to measure? Are you trying to measure the response time of your servers or are you trying to measure the weight of the client in the impact of response time? I put forward the hypothesis that it is possible to measure the client weight by examination of the times captured with the developer tools in both development and also in functional testing.
So much of this client weight is related to page architecture that if you are waiting for performance testing to show you that your page architecture is problematic then you likely will not have time to fix the issues and retest before going to production.
I also recommend the collected O'Reilly works of Steve Souders which will help to bring home the client bound concepts for page design and how much this impacts the end user experience over and above how fast the server responds.
http://www.oreilly.com/pub/au/2951
Related
What other stress test cases are there other than finding out the maximum number of users allowed to login into the web application before it slows down the performance and eventually crashing it?
This question is hard to answer thoroughly since it's too broad.
Anyway many stress tests depend on the type and execution flow of your workload. There's an entire subject dedicated (as a graduate course) to queue theory and resources optimization. Most of the things can be summarized as follows:
if you have a resource (be it a gpu, cpu, memory bank, mechanical or
solid state disk, etc..), it can serve a number of users/requests per
second and takes an X amount of time to complete one unit of work.
Make sure you don't exceed its limits.
Some systems can also be studied with a probabilistic approach (Little's Law is one of the most fundamental rules in these cases)
There are a lot of reasons for load/performance testing, many of which may not be important to your project goals. For example:
- What is the performance of a system at a given load? (load test)
- How many users the system can handle and still meet a specific set of performance goals? (load test)
- How does the performance of a system changes over time under a certain load? (soak test)
- When will the system will crash under increasing load? (stress test)
- How does the system respond to hardware or environment failures? (stress test)
I've got a post on some common motivations for performance testing that may be helpful.
You should also check out your web analytics data and see what people are actually doing.
It's not enough to simply simulate X number of users logging in. Find the scenarios that represent the most common user activities (anywhere between 2 to 20 scenarios).
Also, make sure you're not just hitting your cache on reads. Add some randomness / diversity in the requests.
I've seen stress tests where all the users were requesting the same data which won't give you real world results.
I'm developing a web app that needs to handle bursts of very high loads,
once per minute I get a burst of requests in very few seconds (~1M-3M/sec) and then for the rest of the minute I get nothing,
What's my best strategy to handle as many req /sec as possible at each front server, just sending a reply and storing the request in memory somehow to be processed in the background by the DB writer worker later ?
The aim is to do as less as possible during the burst, and write the requests to the DB ASAP after the burst.
Edit : the order of transactions in not important,
we can lose some transactions but 99% need to be recorded
latency of getting all requests to the DB can be a few seconds after then last request has been received. Lets say not more than 15 seconds
This question is kind of vague. But I'll take a stab at it.
1) You need limits. A simple implementation will open millions of connections to the DB, which will obviously perform badly. At the very least, each connection eats MB of RAM on the DB. Even with connection pooling, each 'thread' could take a lot of RAM to record it's (incoming) state.
If your app server had a limited number of processing threads, you can use HAProxy to "pick up the phone" and buffer the request in a queue for a few seconds until there is a free thread on your app server to handle the request.
In fact, you could just use a web server like nginx to take the request and say "200 OK". Then later, a simple app reads the web log and inserts into DB. This will scale pretty well, although you probably want one thread reading the log and several threads inserting.
2) If your language has coroutines, it may be better to handle the buffering yourself. You should measure the overhead of relying on our language runtime for scheduling.
For example, if each HTTP request is 1K of headers + data, want to parse it and throw away everything but the one or two pieces of data that you actually need (i.e. the DB ID). If you rely on your language coroutines as an 'implicit' queue, it will have 1K buffers for each coroutine while they are being parsed. In some cases, it's more efficient/faster to have a finite number of workers, and manage the queue explicitly. When you have a million things to do, small overheads add up quickly, and the language runtime won't always be optimized for your app.
Also, Go will give you far better control over your memory than Node.js. (Structs are much smaller than objects. The 'overhead' for the Keys to your struct is a compile-time thing for Go, but a run-time thing for Node.js)
3) How do you know it's working? You want to be able to know exactly how you are doing. When you rely on the language co-routines, it's not easy to ask "how many threads of execution do I have and what's the oldest one?" If you make an explicit queue, those questions are much easier to ask. (Imagine a handful of workers putting stuff in the queue, and a handful of workers pulling stuff out. There is a little uncertainty around the edges, but the queue in the middle very explicitly captures your backlog. You can easily calculate things like "drain rate" and "max memory usage" which are very important to knowing how overloaded you are.)
My advice: Go with Go. Long term, Go will be a much better choice. The Go runtime is a bit immature right now, but every release is getting better. Node.js is probably slightly ahead in a few areas (maturity, size of community, libraries, etc.)
How about a channel with a buffer size equal to what the DB writer can handle in 15 seconds? When the request comes in, it is sent on the channel. If the channel is full, give some sort of "System Overloaded" error response.
Then the DB writer reads from the channel and writes to the database.
I need a high performance message bus for my application so I am evaluating performance of ZeroMQ, RabbitMQ and Apache Qpid. To measure the performance, I am running a test program that publishes say 10,000 messages using one of the message queue implementations and running another process in the same machine to consume these 10,000 messages. Then I record time difference between the first message published and the last message received.
Following are the settings I used for the comparison.
RabbitMQ: I used a "fanout" type exchange and a queue with default configuration. I used the RabbitMQ C client library.
ZeroMQ: My publisher publises to tcp://localhost:port1 with ZMQ_PUSH socket, My broker listens on tcp://localhost:port1 and resends the message to tcp://localhost:port2 and my consumer listens on tcp://localhost:port2 using ZMQ_PULL socket. I am using a broker instead of peer to to peer communication in ZeroMQ to to make the performance comparison fair to other message queue implementation that uses brokers.
Qpid C++ message broker: I used a "fanout" type exchange and a queue with default configuration. I used the Qpid C++ client library.
Following is the performance result:
RabbitMQ: it takes about 1 second to receive 10,000 messages.
ZeroMQ: It takes about 15 milli seconds to receive 10,000 messages.
Qpid: It takes about 4 seconds to receive 10,000 messages.
Questions:
Have anyone run similar performance comparison between the message queues? Then I like to compare my results with yours.
Is there any way I could tune RabbitMQ or Qpid to make it performance better?
Note:
The tests were done on a virtual machine with two allocated processor. The result may vary for different hardware, however I am mainly interested in relative performance of the MQ products.
RabbitMQ is probably doing persistence on those messages. I think you need to set the message priority or another option in messages to not do persistence. Performance will improve 10x then. You should expect at least 100K messages/second through an AMQP broker. In OpenAMQ we got performance up to 300K messages/second.
AMQP was designed for speed (e.g. it does not unpack messages in order to route them) but ZeroMQ is simply better designed in key ways. E.g. it removes a hop by connecting nodes without a broker; it does better asynchronous I/O than any of the AMQP client stacks; it does more aggressive message batching. Perhaps 60% of the time spent building ZeroMQ went into performance tuning. It was very hard work. It's not faster by accident.
One thing I'd like to do, but am too busy, is to recreate an AMQP-like broker on top of ZeroMQ. There is a first layer here: http://rfc.zeromq.org/spec:15. The whole stack would work somewhat like RestMS, with transport and semantics separated into two layers. It would provide much the same functionality as AMQP/0.9.1 (and be semantically interoperable) but significantly faster.
Hmm, of course ZeroMQ will be faster, it is designed to be and does not have a lot of the broker based functionality that the other two provide. The ZeroMQ site has a wonderful comparison of broker vs brokerless messaging and drawbacks & advantages of both.
RabbitMQ Blog:
RabbitMQ and 0MQ are focusing on different aspects of messaging. 0MQ puts much more focus on how the messages are transferred over the wire. RabbitMQ, on the other hand, focuses on how messages are stored, filtered and monitored.
(I also like the above RabbitMQ post above as it also talks about using ZeroMQ with RabbitMQ)
So, what I'm trying to say is that you should decide on the tech that best fits your requirements. If the only requirement is speed, ZeroMQ. But if you need other aspects such as persistence of messages, filtering, monitoring, failover, etc well, then that's when u need to start considering RabbitMQ & Qpid.
I am using a broker instead of peer to to peer communication in ZeroMQ to to make the performance comparison fair to other message queue implementation that uses brokers.
Not sure why you want to do that -- if the only thing you care about is performance, there is no need to make the playing field level. If you don't care about persistence, filtering, etc. then why pay the price?
I'm also very leery of running benchmarks on VM's -- there are a lot of extra layers that can affect the results in ways that are not obvious. (Unless you're planning to run the real system on VM's, of course, in which case that is a very valid method).
I've tested c++/qpid
I sent 50000 messages per second between two diferent machines for a long time with no queuing.
I didn't use a fanout, just a simple exchange (non persistent messages)
Are you using persistent messages?
Are you parsing the messages?
I suppose not, since 0MQ doesn't have message structs.
If the broker is mainly idle, you probably haven't configured the prefetch on sender and receptor. This is very important to send many messages.
We have compared RabbitMQ with our SocketPro (http://www.udaparts.com/) persistent message queue at the site http://www.udaparts.com/document/articles/fastsocketpro.htm with all source codes. Here are results we obtained for RabbitMQ:
Same machine enqueue and dequeue:
"Hello world" --
Enqueue: 30000 messages per second;
Dequeue: 7000 messages per second.
Text with 1024 bytes --
Enqueue: 11000 messages per second;
Dequeue: 7000 messages per second.
Text with 10 * 1024 bytes --
Enqueue: 4000 messages per second;
Dequeue: 4000 messages per second.
Cross-machine enqueue and dequeue with network bandwidth 100 mbps:
"Hello world" --
Enqueue: 28000 messages per second;
Dequeue: 1900 messages per second.
Text with 1024 bytes --
Enqueue: 8000 messages per second;
Dequeue: 1000 messages per second.
Text with 10 * 1024 bytes --
Enqueue: 800 messages per second;
Dequeue: 700 messages per second.
Try to configure prefetch on sender and receptor with a value like 100. Prefetching just sender is not enough
We've developed an open source message bus built on top of ZeroMQ - we initially did this to replace Qpid. It's brokerless so it's not a totally fair comparison but it provides the same functionality as brokered solutions.
Our headline performance figure is 140K msgs per second between two machines but you can see more detail here: https://github.com/Abc-Arbitrage/Zebus/wiki/Performance
I want to create a fairly simple mathematical model that describes usage patterns and performance trade-offs in a system.
The system behaves as follows:
clients periodically issue multi-cast packets to a network of hosts
any host that receives the packet, responds with a unicast answer directly
the initiating host caches the responses for some given time period, then discards them
if the cache is full the next time a request is required, data is pulled from the cache not the network
packets are of a fixed size and always contain the same information
hosts are symmetic - any host can issue a request and respond to requests
I want to produce some simple mathematical models (and graphs) that describe the trade-offs available given some changes to the above system:
What happens where you vary the amount of time a host caches responses? How much data does this save? How many calls to the network do you avoid? (clearly depends on activity)
Suppose responses are also multi-cast, and any host that overhears another client's request can cache all the responses it hears - thereby saving itself potentially making a network request - how would this affect the overall state of the system?
Now, this one gets a bit more complicated - each request-response cycle alters the state of one other host in the network, so the more activity the quicker caches become invalid. How do I model the trade off between the number of hosts, the rate of activity, the "dirtyness" of the caches (assuming hosts listen in to other's responses) and how this changes with cache validity period? Not sure where to begin.
I don't really know what sort of mathematical model I need, or how I construct it. Clearly it's easier to just vary two parameters, but particularly with the last one, I've got maybe four variables changing that I want to explore.
Help and advice appreciated.
Investigate tokenised Petri nets. These seem to be an appropriate tool as they:
provide a graphical representation of the models
provide substantial mathematical analysis
have a large body of prior work and underlying analysis
are (relatively) simple mathematical models
seem to be directly tied to your problem in that they deal with constraint dependent networks that pass tokens only under specified conditions
I found a number of references (quality not assessed) by a search on "token Petri net"
I'm evaluating possible solutions for handling a large quantity of queued messages, which must be delivered to workers at a certain date and time. The result of executing them is mostly updates to stored data, and they may or may not be originally triggered by user action.
For example, think of what you'd implement in a hypothetical large-scale StarCraft game server for storing and executing users' actions, like upgrading a building, hatching a soldier, all of which requires to be applied to the game state after several seconds or minutes after the player initiates them.
The problem is I can't seem to find the right term to name this problem area. There are several that looks similar, but different:
cron/task/job scheduler
The content of the queue is not dynamic, it's predefined.
Each task is scheduled.
message queue
The content of the queue is dynamic.
Each task is intended to be delivered immediately.
???
The content of the queue is dynamic.
Each task is scheduled.
If there are message queues that allow conditional delivery of messages, that might be it.
Summary:
What are these kind of technology called?
What are some of the solutions out there?
This just sounds like a trivial priority queue on the surface. The priority in this case is the time of completion, and you check the front of the queue to see when the next event is due. Pretty much every language comes with a priority queue or something that can easily be used as one, so I'm not sure what the actual problem is here.
Is it that you're worried about scalability, when it comes to millions of messages? Obviously 'millions' is a meaningless term - if that's millions per day, it's a trivial problem. If it's millions per second, then you can just scale horizontally, splitting the queue across multiple processes. (And the benefit of such a queue system is that this parallelization is really simple.)
I would bet that when implementing a large scale real-time strategy game server you would hit networking problems long before you start hitting problems with the message queue.
Have you tried looking at push queues by Iron.io? The content of the queue can be anything you like, and you specify a webhook to where the messages will be pushed to. You can also set a delay for each of the messages.
The webhook is static though for each queue and delay isn't always exactly on time (could be up to a minute off). If timing is more important or the ability of providing a different webhook per message is important, try looking at boomerang.io.
They say they are pretty accurate on the timing, you can provide a delay or unix timestamp for the webhook to return and that is per message. Sounds like either of those might work for you.
For StarCraft, I would use the Red Dwarf server.
For a Java EE app, I would use Quartz Scheduler.
It seems to me that a queue-based solution would be best in this case for a number of reasons:
Management. Most queuing solutions provide support for inspecting the content of queues which makes it easier to debug, easier to take action when certain threshold are exceeded, ...
Performance. You can divide workload by having multiple enqueue/dequeue processes (gives you the ability to scale out).
Prioritizing. Most queues support prioritizing of messages (probably not all messages are equally important).
...
Remaining problem is the immediate delivery of messages in the queue. You have two ways to solve this: either delay enqueuing of messages or delay execution of dequeued messages. I would go with the first approach, delayed enqueuing.
A message then has two properties: (content, delay). You provide the message to a component in your system that queues the message at the appropriate time.
I'm not sure what programming language you're using, but the MS .NET 4 framework has support for such a scenario (delayed execution of tasks).