bwlewis
commented
9 years ago Indeed redis sets and lists have an indexing limit of 2^32-1.
You are free to store each element of an R vector in a redis list or set, and then you can store 2^32-1 R objects in a single list or set. But note that these are stored as separate R objects, not elements of a single R vector.
The default behavior of the rredis package serializes R objects (whatever they may be) and stores them as redis "strings" (binary blobs). Using this mechanism, 512MB is the redis-imposed limit of any single R object, vector or otherwise. Again, you are free to use a different convention to get around this redis limit.
Hi
Just a quick question. In the vignette, the size limitation of a R object is indicated below.
When I checked its data types, however, the maximum number of elements of lists, sets and hashes is 2^32 - 1 (http://redis.io/topics/data-types) while the maximum size of its key and value is 512 MB.
So I guess, for example, if a R vector is created, as long as each value (and the key) doesn't exceed 512 MB, it'd be possible to add.
Please inform if the guess is incorrect.
Regards Jaehyeon