Distributed cache

What is a distributed cache what are its benefits?

Distributed caching is a very widely used technique for caching storage data that has been configured across several nodes and servers in the same network. It is an extension of the traditional concept of cache used in a single locale. It is possible for a distributed cache to span multiple servers, which makes it possible for the distributed cache to grow in size and in transactional capacity. Distributed caches are mainly used to store application data residing in database and web session data. It is usually maintained as an external service to the app servers that access it. Instead of using a single cache, it is possible to use more caches in a distributed system to avoid load in the database.

A distributed cache combines the random-access memory (RAM) of several networked computers into a single in-memory data store used as a data cache to provide quick access to data. While the majority of caches are traditionally in a single physical server or hardware component, a distributed cache has the ability to grow beyond the memory limits of a single computer by linking together several computers–referred to as a distributed architecture or a distributed cluster–for higher capacity and greater processing power.

Such caches are particularly useful in environments with high data volume and load. The distributed architecture enables incremental expansion or scaling by adding more computers to the cluster, making it possible for the cache to grow in step with the data growth.

The benefits of using the distributed caching method include:

• You get a reduction in network costs
• You experience an enhancement in responsiveness
• You get your performance optimized on the same hardware settings
• Your content is available 24/7, even when there are network interruptions.

‍

Why is distributed cache faster than database?

Without the use of a cache, when a query is made to the database, the database engine will look up the data, which is generally on disk. Databases even maintain a queue for queries that need to be answered but can’t be processed as yet since the database is busy dealing with other queries. This can result in greater latency response time because of the fact that the web application will have to wait for the database to return the query results.

One method to reduce latency would involve reducing the time that is required to read the data. In some situations, it could even be beneficial to replace hard disk drives with faster SSD drives. But if the volume of queries is high enough that the queue of queries is long even with SSDs, you would have to resort to using an in-memory cache.

When query results are fetched, they get stored in cache. After that, the next time that specific information is required, it gets pulled directly from the cache instead of being pulled from the database. This reduces latency because the data that is required gets fetched from memory, which is faster than disk. It also lowers the number of queries to the database, so queries that can’t be answered by looking up data in the cache won’t have to wait for too long in the query queue before they are processed.

‍

What are the use cases for a distributed cache?

There are several use cases for which an application developer could consider including a distributed cache as part of their architecture. Some of these use cases include:

Application acceleration

Applications that are dependent on disk-based relational databases can’t always match today’s increasingly demanding transaction performance requirements. If you store the most frequently accessed data in a distributed cache, it is possible for you to dramatically reduce the I/O bottleneck of disk-based systems. This helps your applications run much faster, even with a higher volume of transactions when usage spikes.

‍

Storing web session data

A site might store user session data in a cache to serve as inputs for shopping carts and recommendations. With a distributed cache, it is possible to have a large number of concurrent web sessions that can be accessed by any of the web application servers that happen to be running the system. This makes it possible for you to load balance web traffic over multiple application servers and not lose session data incase any application server fails.

‍

Reducing network usage and costs

If you cache data in several places in your network, even on the same computers as your application, it is possible for you to reduce network traffic and leave more bandwidth available for other applications that are dependent on the network.

‍

Minimizing the impact of interruptions

Depending on the architecture, it may be possible for a cache to answer data requests even in situations where the source database is not available.

‍

Extreme scaling

Some applications request particularly high volumes of data. It is possible for a distributed cache to answer those queries by pulling more resources from several machines.

‍

Which technologies make distributed caching possible?

Using a distributed cache has several benefits, but there’s also a rather significant disadvantage to it - the cost of RAM. RAM costs tend to be quite higher than disk or SSD costs, so in-memory speed is not easily available to everyone. It makes sense for businesses that use large distributed caches because they can justify these hardware costs through the quantifiable gains that they receive from having a faster system.

However, RAM costs are currently going down, because of which most businesses can afford in-memory processing. There also are new innovations around memory such as the Intel® Optane™ DC Persistent Memory, which allow businesses to use in-memory solutions such as distributed caches at a far lower cost while leveraging RAM speeds.

There also are distributed cache solutions built on in-memory data grids or in-memory databases.

‍