Week5 - Evolution of Application Data Caching : From RAM to SSD about daily-study HOT 1 CLOSED

Words

• that offers low-latency, high-reliability caching and storage
• in many occasions it is termed as “the hidden microservice”
• performing trillions of operations per day
• has the ability to persist the data to the disk
• With the advent of Netflix global cloud architecture
• The diagram below shows the logical structure of
• As we started moving towards the Global Cloud
• we had a three times increase in the data that needed to be replicated
• Supporting these features came at a considerable increase in cost and complexities
• Our motivation was to provide a global caching solution which was not only fast but was also cost effective.
• are providing fast access to data but at a much lower cost when compared to RAM
• we wanted to move part of the data out of memory without sacrificing availability or performance
• storing the same amount in RAM.
• RAM random read latencies were rarely higher than 1 microsecond
• whereas typical SSD random read speeds are between 100–500 microseconds
• is around 1 millisecond
• we noticed that we were able to perform over
• thus meeting our throughput goals with latency rarely exceeding 1 millisecond
• This meant that
• EVCache Moneta was our first venture at using SSD
• The approach we chose
• This approach reduced the size of most Moneta based clusters over 60% as compared to their corresponding RAM-only clusters
• This enabled us to achieve
• a significant reduction in cost
• However, we were unable to move some of the large online and customer facing clusters
• We would also exceed the desired SLA at times
• As we were working towards solving these issues
• which had taken a different approach
• shim called extstore
• extstore is efficient in terms of cost & storage device utilization
• without compromising the speed and throughput
• All the metadata (key & other metadata) is stored in RAM whereas the actual data is stored on flash.
• which we could not do achieve with Moneta
• we could use at most 50% of the disc capacity
• as we had to ensure an old item could be deleted
• only after it was written again
• This meant we had could end up with a copy of new and old data
• Since we did not have need for storing duplicate records
• are scaled to meet network demands rather than storage demands
• This has been quite a remarkable achievement.
• By moving from
• which allows us to iterate through all of the keys in an instance
• By taking advantage of this command
• at regular intervals
• This ensures data in EVCache is durable and highly available in case of a disaster.
• which is used to store users personalized recommendations.
• The majority of reads are around
• Below is the average read latency of one of cache
• extstore latencies are consistently lower than Moneta for similar load across both the instances.
• we are able to handle all types of workloads
• whether it is a read heavy, write heavy or balanced
• ranging from gigabytes to petabytes
• It has been quite a journey to move from Moneta to extstore
• as of now we have moved all our production clusters running Moneta to extstore
• We have also been able to move
• The new architecture for EVCache Server running extstore
• is allowing us to continue to innovate in ways that matter

from daily-study.