Use of memcache

[cirocosta]

There some intensive/long-running ops which we could optimize using memcache. Below i'll post some docs on it and centralize info.

[cirocosta]

Properties

• key-value,
• in-memory storage,
• never touches the disk,
• 10x faster writes, 5x. faster gets,
• atomic incrementals, also possible w/ compare-and-set for coordinating concurrent access to it.

We could use it to receive any data sent in a pretty frequent way (i.e, counting page events) or for caching frequent query results. Memcache sharding could also be done for very intensive stuff.

GAE also provides the capability of paying for memcache for more reliability/capacity with dedicated memcache. Check it here.

source

• dev.google - Effective memcache
• dev.google - Memcache Python API Overview
• google cloud platform blog - Best Practices for App Engine: memcache and eventual vs strong consistency

[MateusZitelli]

Wow, nice improvement, we must use it! 😃