Rework the filecaching #198
Assignees
Comments
fredrikekre
added a commit
that referenced
this issue
Dec 21, 2023
As described in #198, the file cache is unaware of cache hits for files served directly by nginx. This patch notifies PkgServer.jl about nginx cache hits by sending a mirrored request to the internal /notify endpoint. The original request URI is sent through the `X-Original-URI` header and is used to record the cache hit. In addition, since any cache hits are for files already existing in the cache we can record the size as bytes sent by nginx in a Prometheus counter.
fredrikekre
added a commit
that referenced
this issue
Dec 21, 2023
As described in #198, the file cache is unaware of cache hits for files served directly by nginx. This patch notifies PkgServer.jl about nginx cache hits by sending a mirrored request to the internal /notify endpoint. The original request URI is sent through the `X-Original-URI` header and is used to record the cache hit. In addition, since any cache hits are for files already existing in the cache we can record the size as bytes sent by nginx in a Prometheus counter.
fredrikekre
added a commit
that referenced
this issue
Dec 27, 2023
As described in #198, the file cache is unaware of cache hits for files served directly by nginx. This patch notifies PkgServer.jl about nginx cache hits by sending a mirrored request to the internal /notify endpoint. The original request URI is sent through the `X-Original-URI` header and is used to record the cache hit. In addition, since any cache hits are for files already existing in the cache we can record the size as bytes sent by nginx in a Prometheus counter.
fredrikekre
added a commit
that referenced
this issue
Dec 27, 2023
As described in #198, the file cache is unaware of cache hits for files served directly by nginx. This patch notifies PkgServer.jl about nginx cache hits by sending a mirrored request to the internal /notify endpoint. The original request URI is sent through the `X-Original-URI` header and is used to record the cache hit. In addition, since any cache hits are for files already existing in the cache we can record the size as bytes sent by nginx in a Prometheus counter.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Some initial exploration of the data from #196 suggest that registries have a way to "stick" in the cache. Some datapoints (aggregation over all deployed servers):
In particular it seems wasteful to store so many registries in particular since they are quite large (7x the size of the average package). Registries are also a bit different in that all(?) requests want the latest one, so caching something like the 10 latest should be more than sufficient. This would free up space for packages and artifacts instead, which could potentially reduce network usage between package servers and storage servers.
The major reason causing this (I think) is that we now serve existing files from nginx directly but this cache hit is not recorded on the julia resulting in all resources have
num_accessed = 1.Removing the registries from the equation as suggested above should help quite a bit, but might also be worth trying some way to communicate the cache hits from nginx to julia.
The text was updated successfully, but these errors were encountered: