Is it possible to get a better throughput from our servers, if we make one large HTTP request, as opposed to multiple smaller HTTP requests over HTTP2.
As per my understanding it should not produce any significant difference in performance since with HTTP2 we can have multiple requests multiplexed over one TCP connection.
Yes at a network level one large request will be more efficient than multiple small requests. This is due to the overhead of making a network request.
This is also why concatenating CSS and JavaScript and spriting for images were recommended under HTTP/1.1 so the amount of data sent was the same, but the amount of requests was considerably lower. In fact due to the way compression like gzip works the amount of data was often smaller when sending large requests.
HTTP/2 was designed to make the costs of HTTP requests a lot smaller though the reuse of a single TCP connection using multiplexing. In theory this would allow us to give up concatenation and spriting. The reality has been a little less than perfect though - usually due to browser inefficiencies rather than HTTP/2’s fault. The bottleneck has just moved and we need to optimise browsers for the new world. So, for now, some level of concenation and spriting is still recommended.
Getting back to your question, then yes it should have single effects at that network level and in fact HTTP/1.1 and HTTP/2 may even be similar in performance if you do this.
However beyond the network level you may discover other reasons not to bundle into fewer files. If you have one large JavaScript file for example then the browser must wait for it all to be downloaded before it can be parsed, compiled and run. You may be better getting smaller, more important JavaScript downloaded first. Similarly with image spriting you may be waiting for the entire sprite file to download before a single image is display.
Then there are the caching implications. Changing a single line of JS or adding a single image to the image Sprite requires creating a whole new large file, meaning the older one cannot be used and the whole thing needs to be downloaded again in its entirety.
Plus having large files can be more complicated to implemented and managed. They require a build step (maybe not a big deal as many sites do) and creating and managing image sprites through CSS is often more difficult.
Also if using this to stick with HTTP/1.1, then you may be missing out on are the other benefits of HTTP/2 including HPACK header compression and HTTP/2 push (though this is also more tricky to get right than initially thought/hoped!).
It’s a fascinating topic that I’ve spent a lot of time on, and best advice (as always!) is to understand the technology and test, test, test!
Related
At the company I work, all our APIs send and expect requests/responses that follow the JSON:API standard, making the structure of the request/response content very regular.
Because of this regularity and the fact that we can have hundreds or thousands of records in one request, I think it would be fairly doable and worthwhile to start supporting compressed requests (every record would be something like < 50% of the size of its JSON:API counterpart).
To make a well informed judgement about the viability of this actually being worthwhile, I would have to know more about the relationship between request size and duration, but I cannot find any good resources on this. Anybody care to share their expertise/resources?
Bonus 1: If you were to have request performance issues, would you look at compression as a solution first, second, last?
Bonus 2: How does transmission overhead scale with size? (If I cut the size by 50%, by what percentage will the transmission overhead be cut?)
Request and response compression adds to a time and CPU penalty on both sender's side and receiver's side. The savings in time is in the transmission.
The weighing of the tradeoff depends a lot on the customers of the API -- when they make requests, how much do they request, what is requested, where they are located, type of device/os and capabilities etc.,
If the data is static -- for eg: a REST query apihost/resource/idxx returning a static resource, there are web standard approaches like caching of static resources that clients / proxies will be able to assist with.
If the data is dynamic -- there are architectural patterns that could be used.
If the data is huge -- eg: big scientific data sets, video etc., almost always you would find them being served statically with a metadata service that provides the dynamic layer. For eg: MPEG-DASH or HLS is just a collection of files.
I would choose compression as a last option relative to the other architectural options.
There are also implementation optimizations that would precede using compression of request/response. For eg:
Are your services using all available resources at disposal (cores, memory, i/o)
Does the architecture allow scale-up and scale-out and can the problem be handled effectively using that (remember the penalties on client side due to compression)
Can you use queueing, caching or other mechanisms to make things appear faster?
If you have explored all these and the answer is your system is optimal and you are looking at the most granular unit of service where data volume is an issue, by all means go after compression. Keep in mind that you need to budget compute resources for compression on the server side as well (for a fixed workload).
Your question#2 on transmission overhead vs size is a question around bandwidth and latency. Bandwidth determines how much you can push through the pipe. Latency governs the perceived response times. Whether the payload is 10 bytes or 10MB, latency for a client across the world encountering multiple hops will be larger relative to a client encountering only one or two hops and is bound by the round-trip time. So, a solution may be to distribute the servers and place them closer to your clients from across the world rather than compressing data. That is another reason why compression isn't the first thing to look at.
Baseline your performance and benchmark your experiments for a representative user base.
I think what you are weighing here is going to be the speed of your processor / cpu vs the speed of your network connection.
Network connection can be impacted by things like distance, signal strength, DNS provider, etc; whereas, your computer hardware is only limited by how much power you've put in it.
I'd wager that compressing your data before you are sending would result in shorter response times, yes, but it's=probably going to be a very small amount. If you are sending json, usually text isn't all that large to begin with, so you would probably only see a change in performance at the millisecond level.
If that's what you are looking for, I'd go ahead and implement it, set some timing before and after, and check your results.
Is it possible to gauge a web browsers upload and/or download speed by monitoring normal http requests? Ideally a web application would be able to tell the speed of a client without any modifications and without client-side scripting like JavaScript/Java/Flash. So even if a client was accessing the service with a library like Curl it would still work. If this is possible, how? If its not possible, why? How accurate can this method be?
(If it helps assume PHP/Apache, but really this is a platform independent question. Also being able to gauge the upload speed is more important to me.)
Overview
You're asking for what is commonly called "passive" available bandwidth (ABW) measurement along a path (versus measuring a single link's ABW). There are a number of different techniques1 that estimate bandwidth using passive observation, or low-bandwidth "Active" ABW probing techniques. However, the most common algorithms used in production services are active ABW techniques; they observe packet streams from two different end-points.
I'm most familiar with yaz, which sends packets from one side and measures variation in delay on the other side. The one-sided passive path ABW measurement techniques are considered more experimental; there aren't solid implementations of the algorithms AFAIK.
Discussion
The problem with the task you've asked for is that all non-intrusive2 ABW measurement techniques rely on timing. Sadly, timing is a very tricky thing when working with http...
You have to deal with the reality of object caching (for instance, akamai) and http proxies (which terminate your TCP session prematurely and often spoof the web-server's IP address to the client).
You have to deal with web-hosts which may get intermittently slammed
Finally, active ABW techniques rely on a structured packet stream (wrt packet sizes and timing), unlike what you see in a standard http transfer.
Summary
In summary, unless you set up dedicated client / server / protocol just for ABW measurement, I think you'll be rather frustrated with the results. You can keep your ABW socket connections on TCP/80, but the tools I have seen won't use http3.
Editorial note: My original answer suggested that ABW with http was possible. On further reflection, I changed my mind.
END-NOTES:
---
See Sally Floyd's archive of end-to-end TCP/IP bandwidth estimation tools
The most common intrusive techniques (such as speedtest.net) use a flash or java applet in the browser to send & receive 3-5 parallel TCP streams to each endpoint for 20-30 seconds. Add the streams' average throughput (not including lost packets requiring retransmission) over time, and you get that path's tx and rx ABW. This is obviously pretty disruptive to VoIP calls, or any downloads in progress. Disruptive meausurements are called bulk transfer capacity (BTC). See RFC 3148: A Framework for Defining Empirical Bulk Transfer Capacity Metrics. BTC measurements often use HTTP, but BTC doesn't seem to be what you're after.
That is good, since it removes the risk of in-line caching by denying http caches an object to cache; although some tools (like yaz) are udp-only.
Due to the way TCP connections adapt to available bandwidth, no this is not possible. Requests are small and typically fit within one or two packets. You need a least a dozen full-size packets to get even a coarse bandwidth estimate, since TCP first has to scale up to available bandwidth ("TCP slow start"), and you need to average out jitter effects. If you want any accuracy, you're probably talking hundreds of packets required. That's why upload rate measurement scripts typically transfer several megabytes of data.
OTOH, you might be able to estimate round-trip delay from the three-way handshake and the timing of acks. But download speed has at least as much impact as upload speed.
There's no support in javascript or any browser component to measure upload performance.
The only way I can think of is if you are uploading to a page/http handler, and the page is receiving the incoming bytes, it can measure how many bytes it is receiving per second. Then store that in some application wide dictionary with a session ID.
Then from the browser you can periodically poll the server to get the value in the dictionary using the session ID and show it to user. This way you can tell how's the upload speed.
You can use AJAXOMeter, a JavaScript library which meassures your up- and download speed. You can see a live demo here.
That is not feasible in general as in-bound and out-bound bandwidth frequently is not symmetric. Different ISPs have significantly different ratios here that can vary on even time of the day basis.
I know it's a good practice to minimize the number of requests each page needs. For example, combining javascript files and using css sprites will greatly reduce the number of requests needed to render your page.
Another approach I've seen is to keep javascript embedded in the page itself, especially for javascript specific to that page and not really shared across other pages.
But my question is this:
At what point does my javascript grow too large that it becomes more efficient to pull the script into a separate file and allow the additional request for the separate js file?
In other words, how do I measure how much bytes equates to the cost of one request?
Since successive requests are cached, the only cost of calling that same js file is the cost of the request. Whereas keeping the js in the page will always incur the cost of additional page size, but will not incur the cost of an additional request.
Of course, I know several factors go into this: speed of the client, bandwidth speed, latency. But there has to be a turning point to where it makes more sense to do one over the other.
Or, is bandwidth so cheap (speed, not money) these days that it requires many more bytes than it used to in order to exceed the cost of a request? It seems to be the trend that page size is become less of a factor, while the cost of a request has plateaued.
Thoughts?
If you just look at the numbers and assume an average round-trip time for a request of 100 ms and an average connection speed of 5 Mbps, you can arrive at a number which says that up to 62.5 KB can be added to a page before breaking it out to a separate file becomes worthwhile. Assuming that gzip compression is enabled on your server, then the real amount of JavaScript that you can add is even larger still.
But, this number ignores a number of important considerations. For instance, if you move your JavaScript to a separate file, the user's browser can cache it more effectively such that a user that hits your page 100 times might only download the JavaScript file once. If you don't do this, and assuming that your webpage has any dynamic content whatsoever, then the same user would have to download the entire script every single time.
Another issue to consider is the maintainability of the page. As a general rule, the more JavaScript you add, the more difficult it becomes to maintain your page and make changes and updates without introducing bugs and other problems. So even if you don't have quite 62.5 KB of JavaScript and even if you don't care about the caching side of things, you have to ask yourself whether or not having a separate JavaScript file would improve maintainability and if so, whether it's worth sacrificing that maintainability for a slightly faster page load.
So there really isn't an exact answer here, but as a general rule I think that if the JavaScript is stuff that is truly intrinsic to the page (onclick handlers, effects/animations, other things that interface directly with elements on the page) then it belongs with the page. But if you have a bunch of other code that your handlers, effects, and other things use more like a library/helper utility, then that code can be moved to a separate file. Favor maintainability of your code over both page size and load times. That's my recommendation, anyways.
This is a huge topic - you are indirectly asking about many different aspects of web performance, so there are a few tricks, some of which wevals mentions.
From my own experience, I think it comes down partially to modularization and making tradeoffs. So for instance, it makes sense to pack together javascript that's common across your entire site. If you serve the assets from a CDN and set correct HTTP headers (Cache-Control, Etag, Expires), you can get a big performance boost.
It's true that you will incur the cost of the browser making a request and receiving a 304 Not Modified from the server, but that response at least will be fast to go across the wire. However, you will (typically) still incur the cost of the server processing your request and deciding that the asset is unchanged. This is where web proxies like Squid, Varnish and CDNs in general shine.
On the topic of CDN, especially with respect to JavaScript, it makes sense to pull libraries like jQuery out of one of the public CDNs. For example, Google makes a lot of the most popular libraries available via its CDN which is almost always going to be faster than than you serving it from your own server.
I also agree with wevals that page size is still very important, particularly for international sites. There are many countries where you get charged by how much data you download and so if your site is enormous there's a real benefit to your visitors when you serve them small pages.
But, to really boil it down, I wouldn't worry too much about "byte cost of request" vs "total download size in bytes" - you'd have to be running a really high-traffic website to worry about that stuff. And it's usually not an issue anyway since, once you get to a certain level, you really can't sustain any amount of traffic without a CDN or other caching layer in front of you.
It's funny, but I've noticed that with a lot of performance issues, if you design your code in a sensible and modular way, you will tend to find the natural separations more easily. So, bundle together things that make sense and keep one-offs by themselves as you write.
Hope this helps.
With the correct headers set (far future headers see: 1), pulling the js into a separate file is almost always the best bet since all subsequent requests for the page will not make any request or connection at all for the js file.
The only only exception to this rule is for static websites where it's safe to use a far future header on the actual html page itself, so that it can be cached indefinitely.
As for what byte size equating to the cost of an http connection, this is hard to determine because of the variables that you mentioned as well as many others. HTTP resource requests can be cached at nodes along the way to a user, they can be paralleled in a lot of situations, and a single connection can be reused for multiple request (see: 2).
Page size is still extremely important on the web. Mobile browsers are becoming much more popular and along with that flaky connections through mobile providers. Try and keep file size small.
http://developer.yahoo.com/performance/rules.html
http://en.wikipedia.org/wiki/Hypertext_Transfer_Protocol#Persistent_connections
Addition: It's worth noting that major page size achievements can be achieved through minification and gzip which are super simple to enable through good build tools and web servers respectively.
Has anybody got any experience of using HTTP byte ranges across multiple parallel requests to speed up downloads?
I have an app that needs to download fairly large images from a web service (1MB +) and then send out the modified files (resized and cropped) to the browser. There are many of these images so it is likely that caching will be ineffective - i.e. the cache may well be empty. In this case we are hit by some fairly large latency times whilst waiting for the image to download, 500 m/s +, which is over 60% our app's total response time.
I am wondering if I could speed up the download of these images by using a group of parallel HTTP Range requests, e.g. each thread downloads 100kb of data and the responses are concatenated back into a full file.
Does anybody out there have any experience of this sort of thing? Would the overhead of the extra downloads negate a speed increase or might this actually technique work? The app is written in ruby but experiences / examples from any language would help.
A few specifics about the setup:
There are no bandwidth or connection restrictions on the service (it's owned by my company)
It is difficult to pre-generate all the cropped and resized images, there are millions with lots of potential permutations
It is difficult to host the app on the same hardware as the image disk boxes (political!)
Thanks
I found your post by Googling to see if someone had already written a parallel analogue of wget that does this. It's definitely possible and would be helpful for very large files over a relatively high-latency link: I've gotten >10x improvements in speed with multiple parallel TCP connections.
That said, since your organization runs both the app and the web service, I'm guessing your link is high-bandwidth and low-latency, so I suspect this approach will not help you.
Since you're transferring large numbers of small files (by modern standards), I suspect you are actually getting burned by the connection setup more than by the transfer speeds. You can test this by loading a similar page full of tiny images. In your situation you may want to go serial rather than parallel: see if your HTTP client library has an option to use persistent HTTP connections, so that the three-way handshake is done only once per page or less instead of once per image.
If you end up getting really fanatical about TCP latency, it's also possible to cheat, as certain major web services like to.
(My own problem involves the other end of the TCP performance spectrum, where a long round-trip time is really starting to drag on my bandwidth for multi-TB file transfers, so if you do turn up a parallel HTTP library, I'd love to hear about it. The only tool I found, called "puf", parallelizes by files rather than byteranges. If the above doesn't help you and you really need a parallel transfer tool, likewise get in touch: I may have given up and written it by then.)
I've written the backend and services for the sort of place you're pulling images from. Every site is different so details based on what I did might not apply to what you're trying to do.
Here's my thoughts:
If you have a service agreement with the company you're pulling images from (which you should because you have a fairly high bandwidth need), then preprocess their image catalog and store the thumbnails locally, either as database blobs or as files on disk with a database containing the paths to the files.
Doesn't that service already have the images available as thumbnails? They're not going to send a full-sized image to someone's browser either... unless they're crazy or sadistic and their users are crazy and masochistic. We preprocessed our images into three or four different thumbnail sizes so it would have been trivial to supply what you're trying to do.
If your request is something they expect then they should have an API or at least some resources (programmers) who can help you access the images in the fastest way possible. They should actually have a dedicated host for that purpose.
As a photographer I also need to mention that there could be copyright and/or terms-of-service issues with what you're doing, so make sure you're above board by consulting a lawyer AND the site you're accessing. Don't assume everything is ok, KNOW it is. Copyright laws don't fit the general public's conception of what copyrights are, so involving a lawyer up front can be really educational, plus give you a good feeling you're on solid ground. If you've already talked with one then you know what I'm saying.
I would guess that using any p2p network would be useless as there is more permutations then often used files.
Downloading parallel few parts of file can give improvement only in slow networks (slower then 4-10Mbps).
To get any improvement of using parallel download you need to ensure there will be enough server power. From you current problem (waiting over 500ms for connection) I assume you already have problem with servers:
you should add/improve load-balancing,
you should think of changing server software for something with more performance
And again if 500ms is 60% of total response time then you servers are overloaded, if you think they are not you should search for bottle neck in connections/server performance.
I have reacently activated gzip compression on an IIS6 webserver. I use both static and dynamic compression (static level 10 and dyamic level 1). This was a measure to increase server response time performance. However it seems like the page loads slower after the compression was activated. All my measurements in firebug indicate this.
Has anyone else had this problem? What can be the cause?
You are doing more work on the server and the client so it is normal that response times increase. On low bandwidth connections you might make it good with reduced transfer times.
If you are on a high bandwidth connection then the compression will have not have a significant impact on the transfer delay as it is already short uncompressed. However you will pay 100% of the CPU penalty.
Now zipping large responses takes quite some CPU power, if the server CPU's are already loaded the response times might even get worse.
My advice : check the server CPU and if it is non-negligeable then either turn off zipping or buy a bigger box. If you have a large population on mobile or in remote locations with poor internet connectivity then zipping might be useful, otherwise it will make little difference.
You might also look in using a reverse proxy to reduce the load of the server.
How much bandwidth do you have between your browser and your server?
compressing and decompressing the stream is more work, and so on a fast network, it may actually be slower -- is this an intranet application? You will see the most gain for compression if you have tight bandwidth requirements (either lots of traffic, or a lower bandwidth connection.
How much compression helps will also depend upon what kind of content your site is delivering.
The best thing to do is to test and measure under the same conditions that your site will be under when it is in production.
Static compression works quite well because a copy of the gzipped file is placed in a temporary folder, however dynamically compressed responses have to be re-gzipped every time and unless the bandwidth is a big problem I don't think it is worth it.