Implementing backpressure in the NFS server

2015-02-25 Thread Garrett Wollman
Here's the scenario: 1) A small number of (Linux) clients run a large number of processes (compute jobs) that read large files sequentially out of an NFS filesystem. Each process is reading from a different file. 2) The clients are behind a network bottleneck. 3) The Linux NFS client will

Re: Implementing backpressure in the NFS server

2015-02-25 Thread Alfred Perlstein
On 2/25/15 5:08 PM, Garrett Wollman wrote: Here's the scenario: 1) A small number of (Linux) clients run a large number of processes (compute jobs) that read large files sequentially out of an NFS filesystem. Each process is reading from a different file. 2) The clients are behind a network

Re: Implementing backpressure in the NFS server

2015-02-25 Thread Garrett Wollman
On Wed, 25 Feb 2015 18:29:45 -0500, Alfred Perlstein alf...@freebsd.org said: I think your other suggestions are fine, however the problem is that: 1) they seem complex for an edge case 2) turning them on may tank performance for no good reason if the heuristic is met but we're not in the

Re: Implementing backpressure in the NFS server

2015-02-25 Thread Rick Macklem
Alfred Perlstein wrote: On 2/25/15 5:08 PM, Garrett Wollman wrote: Here's the scenario: 1) A small number of (Linux) clients run a large number of processes (compute jobs) that read large files sequentially out of an NFS filesystem. Each process is reading from a different file.