I have a large Redis sorted set. We need to re-index the data in the set daily, while clients actively request data from the set. My plan is to simply build a second set using a different key and then replace the existing key with the new one:
Build new "indexed" sorted set
RENAME "indexed" set to "live" to replace existing "live" set.
Looking at the RENAME documentation, it states:
If newkey already exists it is overwritten, when this happens RENAME executes an implicit DEL operation, so if the deleted key contains a very big value it may cause high latency even if RENAME itself is usually a constant-time operation.
I'm wondering, then, if it's better to rename the "live" sorted set (e.g. to "dead"), then rename the new "indexed" sorted set to "live" -- and pipeline those requests. And only then, issue a separate DEL command to delete the "dead" set:
Build new "indexed" sorted set
pipeline: RENAME existing "live" set to "dead"
pipeline: RENAME new "indexed" set to "live"
DEL "dead" set
ideas?
Using DEL, you are only postponing the problem. During the DEL, redis blocks other clients.
First, I'd investigate how big the problem is. It can be a problem, for example deleting a 3.5GB ZSET key takes about 2 seconds on our staging system.
If it's a problem, split up the DEL by using ZREMRANGEBYRANK and ZCARD.
Pipelining is efficient (non-transactional ofcourse), so it helps to determine the total size upfront by ZCARD, and after that, issue N ZREMRANGEBYRANK commands (piped) with a range of (example) -10000 0, ending with '0 -1'. As soon as all members are deleted, Redis automatically deletes the key (the Sorted Set) itself.
Hope this helps, TW
Related
We have two separate set of keys in one Redis instance (set1 and set2). All keys in both sets have an expire time set.
If Redis instance hits max memory cap, we want keys from set1 (and only from it!) be evicted to free some memory, but we need to have a guarantee that keys from set2 will not be evicted until their time limit and, thus, will always expire in a normal way.
Is there any possibility to achieve it?
Thanx in advance!
Redis doesn't provide this finely grained of a level of control over cache invalidation. You're restricted to the following options:
noeviction: New values aren’t saved when memory limit is reached. When a database uses replication, this applies to the primary database
allkeys-lru: Keeps most recently used keys; removes least recently used (LRU) keys
allkeys-lfu: Keeps frequently used keys; removes least frequently used (LFU) keys
volatile-lru: Removes least recently used keys with the expire field set to true.
volatile-lfu: Removes least frequently used keys with the expire field set to true.
allkeys-random: Randomly removes keys to make space for the new data added.
volatile-random: Randomly removes keys with expire field set to true.
volatile-ttl: Removes keys with expire field set to true and the shortest remaining time-to-live (TTL) value.
The best you could do would be to set the policy to noeviction and then write your own cache-invalidation process. Or maybe set it to volatile-ttl and then have set2 be non-volatile keys that you remove manually. A fair bit of work and possibly not worth it.
The documentation describing these options also provides some good insight into how Redis actually removes things and might be worth perusing.
Problem:
I need to efficiently delete keys from my Redis Cache using a wildcard pattern. I don't need atomicity; eventual consistency is acceptable.
Tech stack:
.NET 6 (async all the way through)
StackExchange.Redis 2.6.66
Redis Server 6.2.6
I currently have ~500k keys in Redis.
I'm not able to use RedisJSON for various reasons
Example:
I store the following 3 STRING types with keys:
dailynote:getitemsforuser:region:sw:user:123
dailynote:getitemsforuser:region:fl:user:123
dailynote:getitemsforuser:region:sw:user:456
...
where each STRING stores JSON like so:
> dump dailynote:getitemsforuser:region:fl:user:123
"{\"Name\":\"john\",\"Age\":22}"
The original solution used the KeysAsync method to retrieve the list of keys to delete via a wildcard pattern. Since the Redis Server is 6.x, the SCAN feature is being used by KeysAsync internally by the StackExchange.Redis nuget.
Original implementation used a wildcard pattern dailynote:getitemsforuser:region:*. As one would expect, this solution didn't scale well and we started seeing RedisTimeoutExceptions.
I'm aware of the "avoid this in PROD if you can" and have seen Marc Gravell respond to a couple other questions/issues on SO and StackExchange.Redis GitHub. The only potential alternative I could think of is to use a Redis SET to "track" each RedisKey and then retrieve the list of values from the SET (which are the keys I need to remove). Then delete the SET as well as the returned keys.
Potential Solution?:
Create a Redis SET with a key of dailynote:getitemsforuser with a value which is the key of the form dailynote:getitemsforuser:region:XX...
The SET would look like:
dailynote:getitemsforuser (KEY)
dailynote:getitemsforuser:region:sw:user:123 (VALUE)
dailynote:getitemsforuser:region:fl:user:123 (VALUE)
dailynote:getitemsforuser:region:sw:user:456 (VALUE)
...
I would still have each individual STRING type as well:
dailynote:getitemsforuser:region:sw:user:123
dailynote:getitemsforuser:region:fl:user:123
dailynote:getitemsforuser:region:sw:user:456
...
when it is time to do the "wildcard" remove, I get the members of the dailynote:getitemsforuser SET, then call RemoveAsync passing the members of the set as the RedisKey[]. Then call RemoveAsync with the key of the SET (dailynote:getitemsforuser)
I'm looking for feedback on how viable of a solution this is, alternative ideas, gotchas, and suggestions for improvement. TIA
UPDATE
Added my solution I went with below...
The big problem with both KEYS and SCAN with Redis is that they require a complete scan of the massive hash table that stores every Redis key. Even if you use a pattern, it still needs to check each entry in that hash table to see if it matches.
Assuming you are calling SADD when you are also setting the value in your key—and thus avoiding the call to SCAN—this should work. It is worth noting that calls to SMEMBERS to get all the members of a Set can also cause issues if the Set is big. Redis—being single-threaded—will block while all the members are returned. You can mitigate this by using SSCAN instead. StackExchange.Redis might do this already. I'm not sure.
You might also be able to write a Lua script that reads the Set and UNLINKs all the keys atomically. This would reduce network but could tie Redis up if this takes too long.
I ended up using the solution I suggested above where I use a Redis SET with a known/fixed key to "track" each of the necessary keys.
When a key that needs to be tracked is added, I call StackExchange.Redis.IDatabase.SetAddAsync (SADD) while calling StackExchange.Redis.IDatabase.HashSetAsync (HSET) for adding the "tracked" key (along with its TTL).
When it is time to remove the "tracked" key, I first call StackExchange.Redis.IDatabase.SetScanAsync (SSCAN) (with a page size of 250) iterating on the IAsyncEnumerable and call StackExchange.Redis.IDatabase.KeyDeleteAsync (HDEL) on chunks of the members of the SET. I then call StackExchange.Redis.IDatabase.KeyDeleteAsync on the actual key of the SET itself.
Hope this helps someone else.
I have keys in Redis that after being read once are no longer needed. Should I delete them or just let them sit in the database until I need the key again?
I guess the question is which costs more: unneeded data sitting in the database, or a delete operation?
The set command overwrites if string data already exists at the key specified. So, in a way there is a delete and write command, I could use a get and delete command.
Or I could just call delete after getting a key. My question is should I, or just let the key sit there?
If you're only dealing with one key at a time, then overwriting (set) versus deleting (del) both only have a time complexity of O(1) as per the Redis documentation. I'm personally a fan of deleting an entry as soon as I'm finished with it since it is low cost and keeps the storage at a minimum. That being said, both time complexities are low, so overwriting shouldn't be an issue either :)
I can think of two ways of checking existence using redis:
Use the whole database as a 'set', and just SET a key and checking existence by GETing it (or using EXISTS as mentioned in the comment by #Sergio Tulentsev)
Use SADD to add all members to a key and check existence by SISMEMBER
Which one is better? Will it be a problem, compared to the same amount of keys in a single set, if I choose the first method and the number of keys in a database gets larger?
In fact, besides these two methods, you can also use the HASH data structure with HEXISTS command (I'll call this method as the third solution).
All these solutions are fast enough, and it's NOT a problem if you have a large SET, HASH, or keyspace.
So, which one should we use? It depends on lots of things...
Does the key has value?
Keys of both the first and the third solution can have value, while the second solution CANNOT.
So if there's no value for each key, I'd prefer the second solution, i.e. SET solution. Otherwise, you have to use the first or third solution.
Does the value has structure?
If the value is NOT raw string, but a data structure, e.g. LIST, SET. You have to use the first solution, since HASH's value CAN only be raw string.
Do you need to do set operations?
If you need to do intersection, union or diff operations on multiple data sets, you should use the second solution. Redis has built-in commands for these operations, although they might be slow commands.
Memory efficiency consideration
Redis takes more memory-efficient encoding for small SET and HASH. So when you have lots of small data sets, take the second and the third solution can save lots of memory. See this for details.
UPDATE
Do you need to set TTL for these keys?
As #dizzyf points out in the comment, if you need to set TTL for these keys, you have to use the first solution. Because items of HASH and SET DO NOT have expiration property. You can only set TTL for the entire HASH or SET, NOT their elements.
I have a requirement to process multiple records from a queue. But due to some external issues the items may sporadically occur multiple times.
I need to process items only once
What I planned to use is PFADD into redis every record ( as a md5sum) and then see if that returns success. If that shows no increment then the record is a duplicate else process the record.
This seems pretty straightforward , but I am getting too many false positives while using PFADD
Is there a better way to do this ?
Being the probabilistic data structure that it is, Redis' HyperLogLog exhibits 0.81% standard error. You can reduce (but never get rid of) the probability for false positives by using multiple HLLs, each counting a the value of a different hash function on your record.
Also note that if you're using a single HLL there's no real need to hash the record - just PFADD as is.
Alternatively, use a Redis Set to keep all the identifiers/hashes/records and have 100%-accurate membership tests with SISMEMBER. This approach requires more (RAM) resources as you're storing each processed element, but unless your queue is really huge that shouldn't be a problem for a modest Redis instance. To keep memory consumption under control, switch between Sets according to the date and set an expiry on the Set keys (another approach is to use a single Sorted Set and manually remove old items from it by keeping their timestamp in the score).
In general in distributed systems you have to choose between processing items either :
at most once
at least once
Processing something exactly-once would be convenient however this is generally impossible.
That being said there could be acceptable workarounds for your specific use case, and as you suggest storing the items already processed could be an acceptable solution.
Be aware though that PFADD uses HyperLogLog, which is fast and scales but is approximate about the count of the items, so in this case I do not think this is what you want.
However if you are fine with having a small probability of errors, the most appropriate data structure here would be a Bloom filter (as described here for Redis), which can be implemented in a very memory-efficient way.
A simple, efficient, and recommended solution would be to use a simple redis key (for instance a hash) storing a boolean-like value ("0", "1" or "true", "false") for instance with the HSET or SET with the NX option instruction. You could also put it under a namespace if you wish to. It has the added benefit of being able to expire keys also.
It would avoid you to use a set (not the SET command, but rather the SINTER, SUNION commands), which doesn't necessarily work well with Redis cluster if you want to scale to more than one node. SISMEMBER is still fine though (but lacks some features from hashes such as time to live).
If you use a hash, I would also advise you to pick a hash function that has fewer chances of collisions than md5 (a collision means that two different objects end up with the same hash).
An alternative approach to the hash would be to assign an uuid to every item when putting it in the queue (or a squuid if you want to have some time information).