A list of questions I need to ask Martin the next time we meet.
Thesis Layout
[x] Section 2.2 "Related Work", does something like that make sense to include in my thesis?
[x] If yes, what exactly? I see other papers combine this with multi-threaded optimization, sometimes with MIMD, and maybe I can include things like OpenMP etc.?
Answer: Papers that have tested out count-min or AGMS/FAGMS in some setting are good examples.
I remember Martin saying I could stick to some more general information if I don't have enough direct examples, but should also try to keep it brief.
[x] Where do I discuss alternative sketches? - In the Background section
[x] Could Section 2.3 be about background of AGMS and Fast-AGMS?
Input Data
[x] Understand data generation for this thesis (what type of data) 32 bit integers (any)
[x] How do I simulate data streaming for the benchmarks? Feed the algorithm sequentially from a data file? -> Just use the approach like in the sample where you generate data and run through it.
[x] Should I use data sampling for the algorithms? No
Libraries and Implementation
[x] tsimd doesn't support unsigned ints, need to EITHER combine it with the default library to achieve vectorization OR use the additional libsimdpp
[x] (1) "... where m v is the number of members with value v" : What is m in our case? The compare_sketches sample file assigns a static 1.0 as m
[x] I need to vectorize the update functions as well? (EH_3 implementation) Or just whatever makes sense after analysis?
Misc
[x] Get server access to learn using VTune remotely
[x] LaTeX - Where did he gather/use the "List of Tables/Figures/Listings" ? Self-generated, check the main file.
Helpful papers that give insight on what I did in this paper can briefly mentioned.
Note for related work: Fast AGMS is AGMS + Count-Min -> check papers that worked on Count-Min to maybe include some of those insights in the related work.
beepy0
commented
5 years ago 
A list of questions I need to ask Martin the next time we meet.
Thesis Layout
Input Data
Libraries and Implementation
Misc
Helpful papers that give insight on what I did in this paper can briefly mentioned.
Note for related work: Fast AGMS is AGMS + Count-Min -> check papers that worked on Count-Min to maybe include some of those insights in the related work.