Watch Memory Usage Benchmark

NOTE: The watch features are under active development, and their memory usage may change as that development progresses. We do not expect it to significantly increase beyond the figures stated below.

A primary goal of etcd is supporting a very large number of watchers doing a massively large amount of watching. etcd aims to support O(10k) clients, O(100K) watch streams (O(10) streams per client) and O(10M) total watchings (O(100) watching per stream). The memory consumed by each individual watching accounts for the largest portion of etcd's overall usage, and is therefore the focus of current and future optimizations.

Three related components of etcd watch consume physical memory: each grpc.Conn, each watch stream, and each instance of the watching activity. grpc.Conn maintains the actual TCP connection and other gRPC connection state. Each grpc.Conn consumes O(10kb) of memory, and might have multiple watch streams attached.

Each watch stream is an independent HTTP2 connection which consumes another O(10kb) of memory. Multiple watchings might share one watch stream.

Watching is the actual struct that tracks the changes on the key-value store. Each watching should only consume < O(1kb).

                                          | watch |
                              +---------> | foo   |
                              |           +-------+
                       |   stream   |
      +--------------> |            |
      |                +------+-----+     +-------+
      |                       |           | watch |
      |                       +---------> | bar   |
+-----+------+                            +-------+
|            |         +------------+
|   conn     +-------> |   stream   |
|            |         |            |
+-----+------+         +------------+
      |                +------------+
      +--------------> |   stream   |
                       |            |

The theoretical memory consumption of watch can be approximated with the formula: memory = c1 * number_of_conn + c2 * avg_number_of_stream_per_conn + c3 * avg_number_of_watch_stream

Testing Environment

etcd version

  • git head

GCE n1-standard-2 machine type

  • 7.5 GB memory
  • 2x CPUs

Overall memory usage

The overall memory usage captures how much RSS etcd consumes with the client watchers. While the result may vary by as much as 10%, it is still meaningful, since the goal is to learn about the rough memory usage and the pattern of allocations.

With the benchmark result, we can calculate roughly that c1 = 17kb, c2 = 18kb and c3 = 350bytes. So each additional client connection consumes 17kb of memory and each additional stream consumes 18kb of memory, and each additional watching only cause 350bytes. A single etcd server can maintain millions of watchings with a few GB of memory in normal case.

clients streams per client watchings per stream total watching memory usage
1k 1 1 1k 50MB
2k 1 1 2k 90MB
5k 1 1 5k 200MB
1k 10 1 10k 217MB
2k 10 1 20k 417MB
5k 10 1 50k 980MB
1k 50 1 50k 1001MB
2k 50 1 100k 1960MB
5k 50 1 250k 4700MB
1k 50 10 500k 1171MB
2k 50 10 1M 2371MB
5k 50 10 2.5M 5710MB
1k 50 100 5M 2380MB
2k 50 100 10M 4672MB
5k 50 100 50M OOM