Invalidating the cache

28-Dec-2019 17:33 by 8 Comments

Invalidating the cache

This kind of positive feedback loops are a recipe for disaster.I do not see a solution to high sensitivity of the distributed cache to load spikes and network latency spikes.

invalidating the cache-46

For large organizations with process overhead in generating tickets, invalidating cache can take time. So the request has come in quite frequently..."How can I invalidate cache remotely?

Also, last thing I need when my server is loaded is to start repair and data shuffling.

It is possible to run a distributed cache without trouble if you do lot of monitoring and hand holding to make sure nodes are not loaded.

That is network partition the nodes to two clusters.

Then system has to make sure only one half is working ( otherwise, you might end up with conflicting updates and stale data).

In this case, the only data I need to pass is to set a key.

To set an i Stats key, you need to specify : I'm not measuring anything, so I'll use a string starting with "WA policy string" and followed by the name of the policy.

With i Stats-based triggers, you need linkage to bind the i Stats key to an event-name, wacache in my case.

You can also set thresholds and durations, but again since I am not measuring anything, that isn't necessary.

You can be explicit or allow the users to pass it in a query parameter as I'm doing in this i Rule below: Setting the key this way will allow you to create as many triggers as you have policies.

I'll leave it as an exercise for the reader to make that step more dynamic.

Most our servers are stateless ( that is they keep their state in a data base). ( there are several ways to do this, and they usually involve heartbeat and a consensus algorithm). The presumed failed node does not know it has failed and continue to work. When it joins back, system need to reshuffle the data again.