Increase default 'max_semi_space_size' value to reduce GC overhead in V8

[JialuZhang-intel]

What is the problem this feature will solve?

When I use node to run the web-tooling-benchmark, I found that the runtime flag --max_semi_space_size have a big impact on the test result. The total throughput increased about 18% after I pass the runtime flag --max_semi_space_size=128 into node. So I did some investigate about the 'max_semi_space_size' flag.

From some v8 official blogs (Getting garbage collection for free, orinoco-parallel-scavenger), I found there are two garbage collection strategies in V8:

Scavenge GC: fast, high frequency, for young generation objects.

Major GC: take a long time, low frequency, full garbage collections.

When we create a new object by javascript code, the object will be put into semi_space as a young generation object. And when the semi_space is about to use up, V8 engine will trigger the Scavenge GC to clean up the garbage objects in semi_space.

If I use the --max_semi_space_size flag to increase the maximum limit of semi_space size, the scavenge GC occur frequency will decrease. This will bring both advantages and disadvantages:

• Advantage: throughput improvement. Because the scavenge GC occur frequency decreased, the total GC pause time will also decrease, then node can have more CPU resource to execute the javascript code.
• Disadvantage: more memory usage. This is obviously, more semi_space size will cost more memory.

It's a trade-off between time and space. V8 set the default max_semi_space_size as 16MB for 64bit system and 8MB for 32bit system (related code). I think it's a heuristic value that mainly considered client device with small RAM size (for example: some android device only have 4GB RMA). But for server scenarios, memory usually isn't the bottleneck, while throughput is the actual bottleneck.

So the problem is:

Whether the currently default max_semi_space_size(16MB/8MB) for V8 is also the best configuration for node?

What is the feature you are proposing to solve the problem?

To solve the problem above, I tuned the --max_semi_space_size (16MB, 32MB, 64MB, 128MB, 256MB) and tested on web-tooling-benchmark and a simple service based on ghost.js. Here is the test results:

From the figure we can see that:

• Peak memory usage increases linearly with max_semi_space_size.
• Throughput grows fast when max_semi_space_size less than 128MB, then keep flat when max_space_size bigger than 128MB.
• The scale of the throughput improvement is workload-dependent, probably due to the greater GC pressure from web-tooling-benchmark.

So I think we can choose a better max_semi_space_size value and pass this runtime flag to V8 when node startup.

What alternatives have you considered?

Test environment:

• Hardware:
  • CPU: Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz
  • RAM: 500GB
• Software:
  • OS: Ubuntu 20.04.1 (x86_64)
  • Linux version: 5.11.0-41-generic
  • Docker version: 20.10.11
  • node version: v18.0.0-pre

Test process:

1. Build the docker container with node binary and workload in it.
2. Start multi-containers (containers number equals vCPU number) to make sure the system's total CPU usage rate >90%.
3. Running the workload in started containers concurrently and monitor the system's total memory usage periodically.

[targos]

@nodejs/v8

[joyeecheung]

I remember there were similar requests for max_old_space_size (whose relatively small value is the source of a lot of confusion) - if we decide to use different defaults, I suppose these should all be tweaked to an optimized setting, though it's difficult to tell which setting is the best as a one-size-fits-all solution and I think that's why we've been keeping the V8 defaults so far (and another factor is that these are all configurable via the command line and in general Node.js core prefers to let the user choose what's best for their use cases instead of picking one itself)

[JialuZhang-intel]

@joyeecheung Thanks for your explanation! Is there any official documents about the Node.js core prefers? And If there are
some documents, can we put this preferable max_semi_space_size configuration in it?

[ronag]

I guess this also applies to workers and the maxYoungGenerationSizeMb option?

[ronag]

Actually it doesn't seem to be possible to set semi space for workers? @jasnell @addaleax

Only:

• max_young
• max_old

Are possible to set. https://lbwa.github.io/v8-reference/classv8_1_1_resource_constraints.html

[ronag]

and another factor is that these are all configurable via the command line and in general Node.js core prefers to let the user choose what's best for their use cases instead of picking one itself

The problem with this is that 99% of users don't know what is best for their use case so I guess we have to make some kind of decision in terms of defaults that works best for most of our users. I guess the current defaults are more optimized for browser workloads rather than server workloads? IMHO it might be worth looking into optimizing the defaults to better suit our users.

[joyeecheung]

And If there are some documents, can we put this preferable max_semi_space_size configuration in it?

We already document --max-old-space-size with some suggestions (https://github.com/nodejs/node/blob/master/doc/api/cli.md#--max-old-space-sizesize-in-megabytes), so adding documentation for the semi space size there makes sense, too.

[JialuZhang-intel]

@ronag

The problem with this is that 99% of users don't know what is best for their use case.

I agree with this, and V8 should consider the browser's memory consumption in mobile device with small RAM size. But for server scenarios, memory usually isn't the bottleneck.

V8 has provided the interface set_max_semi_space_size_in_kb() to change the default max_semi_space_size. I found a related issue try to setup the default max_yong and max_old generation size according to system's physical_memory, which uses the ResourceConstraints::ConfigureDefaults interface in V8. But V8 has limited the max heap size as 2GB, and there is a proportional relationship between semi_space_size and heap_size, so the default max_semi_space_size can't large than 16MB. Can we create a similar ConfigureDefaults function for node and choose an optimal configuration for server scenarios?

[JialuZhang-intel]

@joyeecheung

We already document --max-old-space-size with some suggestions (https://github.com/nodejs/node/blob/master/doc/api/cli.md#--max-old-space-sizesize-in-megabytes), so adding documentation for the semi space size there makes sense, too.

OK, I added the --max_semi_space_size introduction into the document, this is the related PR (#42575).