mbox series

[ovs-dev,v1,0/9] conntrack: improve multithread scalability

Message ID cover.1613557616.git.grive@u256.net
Headers show
Series conntrack: improve multithread scalability | expand

Message

Gaƫtan Rivet Feb. 17, 2021, 4:34 p.m. UTC
Conntracks are executed within the datapath. Locks along this path are crucial
and their critical section should be minimal. The global 'ct_lock' is necessary
before any action taken on connection states. This lock is needed for many
operations on the conntrack, slowing down the datapath.

The cleanup thread 'ct_clean' will take it to do its job. As it can hold it a
long time, the thread is limited in amount of connection cleaned per round,
and calls are rate-limited.

* Timeout policies locking is contrived to avoid deadlock.
  Anytime a connection state is updated, during its update it is unlocked,
  'ct_lock' is taken, then the connection is locked again. Then the reverse
  is done for unlock.

* Scalability is poor. The global ct_lock needs to be taken before applying
  any change to a conn object. This is backward: local changes to smaller
  objects should be independent, then the global lock should only be taken once
  the rest of the work is done, the goal being to have the smallest possible
  critical section.

It can be improved. Using RCU-friendly structures for connections, zone limits
and timeout policies, read-first workload is improved and the precedence of the
global 'ct_lock' and local 'conn->lock' can be inversed.

Running the conntrack benchmark we see these changes:
  ./tests/ovstest test-conntrack benchmark <N> 3000000 32

code \ N      1     2     4     8
  Before   2310  2766  6117 19838  (ms)
   After   2072  2084  2653  4541  (ms)

One thread in the benchmark executes the task of a PMD, while the 'ct_clean' thread
runs in background as well.

Github actions: https://github.com/grivet/ovs/actions/runs/574446345

Gaetan Rivet (9):
  conntrack: Use rcu-lists to store conn expirations
  conntrack: Use a cmap to store zone limits
  conntrack: Init hash basis first at creation
  conntrack-tp: Use a cmap to store timeout policies
  conntrack: Inverse conn and ct lock precedence
  conntrack: Do not schedule zero ms timers
  conntrack: Do not rate limit ct-sweep
  conntrack: Do not log empty ct-sweep
  conntrack: Use an atomic conn expiration value

 lib/conntrack-private.h |  96 +++++++++++++------
 lib/conntrack-tp.c      | 137 +++++++++++++++-----------
 lib/conntrack.c         | 206 +++++++++++++++++++++++++---------------
 lib/conntrack.h         |   4 +-
 lib/dpif-netdev.c       |   5 +-
 5 files changed, 280 insertions(+), 168 deletions(-)

--
2.30.0