On Mon, Jan 27, 2020 at 09:34:25AM -0800, Roman Gushchin wrote: > The existing cgroup slab memory controller is based on the idea of > replicating slab allocator internals for each memory cgroup. > This approach promises a low memory overhead (one pointer per page), > and isn't adding too much code on hot allocation and release paths. > But is has a very serious flaw: it leads to a low slab utilization. > > Using a drgn* script I've got an estimation of slab utilization on > a number of machines running different production workloads. In most > cases it was between 45% and 65%, and the best number I've seen was > around 85%. Turning kmem accounting off brings it to high 90s. Also > it brings back 30-50% of slab memory. It means that the real price > of the existing slab memory controller is way bigger than a pointer > per page. > > The real reason why the existing design leads to a low slab utilization > is simple: slab pages are used exclusively by one memory cgroup. > If there are only few allocations of certain size made by a cgroup, > or if some active objects (e.g. dentries) are left after the cgroup is > deleted, or the cgroup contains a single-threaded application which is > barely allocating any kernel objects, but does it every time on a new CPU: > in all these cases the resulting slab utilization is very low. > If kmem accounting is off, the kernel is able to use free space > on slab pages for other allocations. > > Arguably it wasn't an issue back to days when the kmem controller was > introduced and was an opt-in feature, which had to be turned on > individually for each memory cgroup. But now it's turned on by default > on both cgroup v1 and v2. And modern systemd-based systems tend to > create a large number of cgroups. > > This patchset provides a new implementation of the slab memory controller, > which aims to reach a much better slab utilization by sharing slab pages > between multiple memory cgroups. Below is the short description of the new > design (more details in commit messages). > > Accounting is performed per-object instead of per-page. Slab-related > vmstat counters are converted to bytes. Charging is performed on page-basis, > with rounding up and remembering leftovers. > > Memcg ownership data is stored in a per-slab-page vector: for each slab page > a vector of corresponding size is allocated. To keep slab memory reparenting > working, instead of saving a pointer to the memory cgroup directly an > intermediate object is used. It's simply a pointer to a memcg (which can be > easily changed to the parent) with a built-in reference counter. This scheme > allows to reparent all allocated objects without walking them over and > changing memcg pointer to the parent. > > Instead of creating an individual set of kmem_caches for each memory cgroup, > two global sets are used: the root set for non-accounted and root-cgroup > allocations and the second set for all other allocations. This allows to > simplify the lifetime management of individual kmem_caches: they are > destroyed with root counterparts. It allows to remove a good amount of code > and make things generally simpler. > > The patchset* has been tested on a number of different workloads in our > production. In all cases it saved significant amount of memory, measured > from high hundreds of MBs to single GBs per host. On average, the size > of slab memory has been reduced by 35-45%. Here are some numbers from multiple runs of sysbench and kernel compilation with this patchset on a 10 core POWER8 host: ========================================================================== Peak usage of memory.kmem.usage_in_bytes, memory.usage_in_bytes and meminfo:Slab for Sysbench oltp_read_write with mysqld running as part of a mem cgroup (Sampling every 5s) -------------------------------------------------------------------------- 5.5.0-rc7-mm1 +slab patch %reduction -------------------------------------------------------------------------- memory.kmem.usage_in_bytes 15859712 4456448 72 memory.usage_in_bytes 337510400 335806464 .5 Slab: (kB) 814336 607296 25 memory.kmem.usage_in_bytes 16187392 4653056 71 memory.usage_in_bytes 318832640 300154880 5 Slab: (kB) 789888 559744 29 -------------------------------------------------------------------------- Peak usage of memory.kmem.usage_in_bytes, memory.usage_in_bytes and meminfo:Slab for kernel compilation (make -s -j64) Compilation was done from bash that is in a memory cgroup. (Sampling every 5s) -------------------------------------------------------------------------- 5.5.0-rc7-mm1 +slab patch %reduction -------------------------------------------------------------------------- memory.kmem.usage_in_bytes 338493440 231931904 31 memory.usage_in_bytes 7368015872 6275923968 15 Slab: (kB) 1139072 785408 31 memory.kmem.usage_in_bytes 341835776 236453888 30 memory.usage_in_bytes 6540427264 6072893440 7 Slab: (kB) 1074304 761280 29 memory.kmem.usage_in_bytes 340525056 233570304 31 memory.usage_in_bytes 6406209536 6177357824 3 Slab: (kB) 1244288 739712 40 -------------------------------------------------------------------------- Slab consumption right after boot -------------------------------------------------------------------------- 5.5.0-rc7-mm1 +slab patch %reduction -------------------------------------------------------------------------- Slab: (kB) 821888 583424 29 ========================================================================== Summary: With sysbench and kernel compilation, memory.kmem.usage_in_bytes shows around 70% and 30% reduction consistently. Didn't see consistent reduction of memory.usage_in_bytes with sysbench and kernel compilation. Slab usage (from /proc/meminfo) shows consistent 30% reduction and the same is seen right after boot too. Regards, Bharata.
