1,749 reputation
818
bio website stackoverflow.com
location New York, NY
age 42
visits member for 3 years, 11 months
seen 20 hours ago

StackOveflow profile

Areas of strength:

  • Perl expert (specifically enterprise software development)
  • Sybase (including design and optimization)
  • GUI design

Areas of familiarity:

  • Web programming (EmbPerl, JSP, CSS, HTML, JavaScript)
  • C++

...P.S. I'm not really 42. But 42 is way cooler than a real answer :)

...P.P.S. Don't read too much into the icon. Just a minor nod to Cryptonomicon.


Oct
28
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
that's why when the question asks for data structures and algorithms, it asks for just that, and not specific implementation :)
Oct
27
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
Using Windows filesystem for a unix server would be a cute trick
Oct
20
comment Data structures and algorithms for a directed rooted tree with inherited properties?
@btilly - yes, that's definitely the kind of architecture I will be implementing. However, the lag from changes to reflect in the updated caches should be minimal so algorithms and performance would still matter a lot.
Oct
20
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
@RobertHarvey - seems like an accurate assessment. I'm not saying a great solution exists, I'm saying I don't know enough about advanced tree algorithms to know for sure and hoping that some meaningful optimizations can be made, even if they aren't of magic wand variety. Hell I'll take linear optimizations :)
Oct
20
revised Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
edited body
Oct
20
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
@RobertHarvey - I'm not too familiar with that structure beyond skimming the Wiki article now, but how would it help here? I would think it would actually be more useful to my other question (because it does contain "is X in a set" questions on data that has a fairly large "standard" distribution and a smaller set of outliers)
Oct
20
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
@RobertHarvey - It might speed things up on retrieval of course, linearly. Question is what impact would it have on tree changes when you need to re-compute those caches (and know WHICH caches to recompute based on which tree nodes changed). That is precisely the sort of reasoning I was hoping to get in an answer :)
Oct
20
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
@RobertHarvey - I thought I covered all of them except "store 5 at a time" and frankly I'm uncertain what you meant by that. Cache 5 immediate ancestors?
Oct
20
revised Data structures and algorithms for a directed rooted tree with inherited properties?
added 484 characters in body
Oct
20
comment Data structures and algorithms for a directed rooted tree with inherited properties?
@msw - access patterns added.
Oct
20
comment Data structures and algorithms for a directed rooted tree with inherited properties?
@msw - Data structure is a directed rooted tree. Implementation is wide open - I'm redesigning the software implementation from scratch and as such, would be open to ANY structure proposed.
Oct
20
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
@RobertHarvey - same problem domain, but would likely end up with different data structures (from past experience in the problem domain, the solution to the other one would strongly optimize towards the properties speed and be less optimal to pure tree side of it). It's possible that some approaches from this question may be helpful to solving another one however.
Oct
20
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
@ChrisF - this is NOT a duplicate. There are different tradeoffs and different data structures and algorithms for both cases (an ideal solution to the other question MAY answer this one, but IMHO due to tradeoffs that ideal solution would not exist). As such, an algorithm that ideally solves the linked question would most likely be sub-optimal for the tree one asked here due to optimizing for properties actions and not tree actions.
Oct
20
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
@RobertHarvey - I'm pretty language agnostic. Realistically would likely be Java or Perl or Python (current implementation is Sybase T-SQL, of all the fun things to maintain :))))
Oct
20
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
@RobertHarvey - yep, that's the choices I tried and both suck. I was trying to see if there's a more efficient, less obvious approach that smoothes the tradeoffs.
Oct
20
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
@RobertHarvey - the tree can get ~50 deep - and it's not terribly balanced. Also, finding a PATH implies concatenation, doesn't it?
Oct
20
asked Data structures and algorithms for a directed rooted tree with inherited properties?
Oct
20
comment Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
@RobertHarvey - how can you do a less than O(D) path-up for a node in a regular tree structure? Wouldn't you have to cache the paths in some way? I agree that adding a hashtable would be one way to solve the name lookup (that's how I'm doing it now).
Oct
20
asked Which data structures and algorithms should I consider for a directed rooted tree that has high churn?
Sep
30
awarded  Explainer