back to article AWS makes its own Arm CPUs the default for ElastiCache in-memory data store service

Amazon Web Services has made its home-brewed Arm-powered Graviton2 processors the default for its ElastiCache service. ElastiCache is AWS’s in-memory-data-store-as-a-service and lets users create Redis or memcached implementations in the Amazonian cloud. AWS got all excited about ElastiCache on Graviton2 back in December 2019 …

  1. ForthIsNotDead
    Thumb Up


    I don't see the 'code not cut for Arm" thing being much of an issue. If, as the article mentions, you are using Redis on ElastiCache then your in-memory DB is merely a service behind an API. You wouldn't really know or care what silicon is hosting it. As it should be. If Amazon are truly willing to pass on cost savings to the end-user then well done to them.

  2. Aitor 1 Silver badge


    Ahh nothing like rich hostages, I mean "customers".

    This is why monopolies and oligopolies are bad: they force you on these things.. so now they want to force feed their own silicon.. and as they control the pricing, they make it cheaper...

    1. PC Master Race

      Re: Monopolies.,..

      You must be joking? They are changing the landscape for the customer with CHEAPER and FASTER silicon. Yes, they control it. But also Intel and AMD control their silicon.. In my world - Intel needs to step up..

  3. aussie-alan

    Changing to ARM should be easy

    The last couple of startups I wrote code for did all their work in Python, NodeJS or Java - it should make no difference going to ARM in x86. If they move services such as RDS / Aurora / whatever, I shouldn't really care about the platform, only the price / performance. Now, if you write C/C++ code and depend on endianness, size of integers, etc, than a) it's most likely poor code, and b) it's your problem. There are exceptions, but experienced coders will define their assumptions and differences in a couple of header files and libraries, then write platform-independent code.

    BTW, I code on a Mac, deploy on Linux (several different clouds), as well as deploy Python into Windows for one of my customers. That's a more significant difference than worrying about the CPU architecture.


POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2022