1
6
2
8
3
0
4
14
5
4

ULL trading firms go to a lot of trouble to get their servers and switches within the same buildings as the exchanges they trade with to reduce latency. Some firms don’t even use layer 1 switches to be competitive.

6
6
7
24
8
17

My kernels go 2x faster than MKL for matrices that fit in L2 cache, which makes them a work in progress, since the speedup works best for prompts having fewer than 1,000 tokens.

9
6
10
2
Trace Compass (eclipse.dev)
11
20
12
3
13
2
eBPF Documentary (www.brendangregg.com)
14
10
Perf is not enough (motherduck.com)
15
3
16
8
UUID Benchmark War (ardentperf.com)
17
6
18
14
19
3
20
5
21
3
22
4
Apache SkyWalking (skywalking.apache.org)
23
4
24
7
25
1
view more: next ›

Performance

286 readers
1 users here now

A community for posts relating to performance

Wormhole

!programming@programming.dev

founded 11 months ago
MODERATORS