You didn't say which language you're using, but if you're on Java, I strongly suggest Objectify.
It defaults all columns/fields to unindexed unless you decorate the field with an @Index annotation. On Wednesday, March 5, 2014 2:55:12 AM UTC-6, Mattan Furst wrote: > > I have a question regarding performance and saving non-indexed data in the > datastore. > > I have a situation where I have an entity with several tidbits of > information that I need to save (currently around 20-30 columns each with a > few bytes of information worth). Most of those tidbits do not require > indexing. It used to be that I would save every piece of information in a > different indexed column. After reading up on the subjects I came to > realize that I can save costs and performance by not having all the pieces > of information in indexed column. I have 3 main options I consider changing > into: > 1. Keep saving the various column I don't need indexing for but turn their > type to ByteString to keep them from being indexed. > 2. Get all the tidbits of information, turns them into some common format > like json or xml and store that in a single blob column. > 3. Same thing as 2 but before storing the blob compress it using some fast > and common compression algorithm like. > > Which is the better option? > What are the considerations I need to make as far as write read costs or > cpu time? > Can you direct me to reading material on the subject (performance and data > storage) specifically tuned to google's datastore? > -- You received this message because you are subscribed to the Google Groups "Google App Engine" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/google-appengine. For more options, visit https://groups.google.com/d/optout.
