amitbhadra / twitterapp Goto Github PK

The members of this team - Amit Asish Bhadra - 66494087 Rishab Aryan Das - 05369273

1. Optimum work unit is choosing 12 number of actors because my number of cores is 12 and choosing anything more or less causes the concurrency to decrease. This ratio is calculated over many number of actors and the best possible one is 12 for me. I have noticed that when the input size is too low, I'm getting low concurrency, the beauty of parallelism shows only when the input N and k is increased a lot.

2. Please run the code as - dotnet fsi --langversion:preview program.fsx There is no result for 10^6 and 4.

However, the result for 10^8 and 24 is as below -- 9 1 20 44 76 25 304 197 121 353 540 1301 2053 3597 856 3112 5448 8576 12981 20425 30908 35709 54032 84996 128601 202289 306060 353585 534964 841476 1273121 2002557 3029784 3500233 5295700 8329856 12602701 19823373 22579140 58980291 82457176 90916537 Real: 00:02:53.857, CPU: 00:21:52.822, GC gen0: 30363, gen1: 7383, gen2: 18

For 10^6 and 24 it is -- 1 9 20 44 197 353 25 121 76 304 540 1301 856 2053 3112 3597 5448 8576 12981 20425 30908 35709 54032 84996 128601 202289 306060 353585 534964 841476 Real: 00:00:01.177, CPU: 00:00:12.248, GC gen0: 303, gen1: 70, gen2: 1

For 10^8 and 20 it is -- 53553387 62780852 88700958 Real: 00:02:42.922, CPU: 00:21:25.719, GC gen0: 28302, gen1: 6837, gen2: 7

3. The maximum concurrency I have got is 10.4. For values where number of actors are around 500-1000, I'm getting a concurrency of 6-7

4. The largest problem I managed to solve is 10^8