Redis Memory Optimization
This quick session covers how effective and handy compression can be while storing heavy objects in Redis. The majority of our products and offerings rely heavily on Redis for effective caching and response retrieval. TWe have had issues with running out of memory, having to failover, having to vertically upscale, and needing to add more memory. We have also faced running out of space, and needing to quickly clear off less impactful keys and values in order not to degrade production performance. We recently explored Snappy and LZ4 compression techniques prior to having Redis store these values. With the Redis solution, we foresee a 46% reduction in memory consumption and network bytes transferred. We will share our use case, performance testing results, and present metrics.