Reputation
8,643
Top tag
Next privilege 10,000 Rep.
Access moderator tools
Badges
1 11 21
Newest
 Guru
Impact
~416k people reached

Aug
18
comment Using a distributed system as an alternative to disk storage
I think you'd be better off with a RAID array of SSDs, unless you're working with Infiniband or some other insanely fast network. And once you start dispersing your network, latency creeps in and will kick your throughput in the gnuts.
Jul
9
awarded  Guru
Jul
2
answered How do binary trees use memory to store its data?
Jun
3
answered Architecture: Should I schema validate incoming reply messages?
May
29
answered How to use one web page to create and edit database entries
May
21
answered Should I use BLOB or Tables for storing large data?
May
11
answered Structuring Procedural vs OO code
Apr
28
answered Is it better to make database calls or external API calls first in the context of a single Web request?
Mar
12
answered Working with two different database types for one application
Feb
2
comment S3 image urls in database
@the_: UUIDs would be better, since random number generators can (by design) produce the same number more than once. Since you're using Postgres, I'd recommend just using a sequence.
Jan
30
comment S3 image urls in database
@the_: Use something like a UUID or a database sequence (if your database software offers one). If your data volume is low enough, you could probably get away with something like (user ID + timestamp).
Jan
29
comment S3 image urls in database
Three things you can do: (1) save the parent first to get the ID, (2) have an ID generator you can call to set the ID before you save it, or (3) generate a separate (non-primary) ID to act as a surrogate key and store that in a separate field in Product (say ProductImageID).
Jan
20
answered Scaling locks in high concurrency web apps
Jan
15
comment Is there any material difference between queries joined by WHERE clauses, and queries using an actual JOIN?
The reason the JOIN statement was added was because there was no standard for expressing outer joins, so each database vendor had their own "special" (incompatible) syntax for it. IIRC Oracle had *= or =* indicating left or right outer joins, another one I used only supported left outer joins using an |= operator.
Jan
15
answered Saving to file before storing into a database in a REST service
Jan
15
answered Is there any material difference between queries joined by WHERE clauses, and queries using an actual JOIN?
Jan
15
comment How to process large files in NetLogo?
Have you tried profiling your code to see where the hotspots are? Sometimes fairly simple changes (e.g., reading a batch of lines instead of reading them individually) can improve performance significantly.
Jan
15
comment Relation between cache line and memory page
The answers to your questions are highly CPU-architecture-dependent.
Jan
14
comment What's the best way to synchronise an event over multiple processes?
Sure it does -- it syncs on every block boundary. Clients aren't allowed to start the next block until both have signaled that they have completed playing the previous block. It would certainly be possible to have both clients preload the entire content before playback, but I saw that as an implementation detail best determined by the available network capabilities.
Jan
13
answered What's the best way to synchronise an event over multiple processes?