Introduction:
Salesforce Hyperforce is the next generation of Salesforce multi-tenancy architecture, allowing Salesforce clients to store their data in public cloud storage centers such as Amazon Web Services (AWS), Google Cloud, and Microsoft Azure rather than in Salesforce data centers. Whereby offering clients more choice and control over data residency.
What makes hyperforce special?
- Immutable Infrastructure: Salesforce has run services on a fixed set of servers in data centers which they own, since its start in 1999. Changes on these hosts like OS and service upgrades are managed by operators using different tools, where these tools are based on mutable infrastructure i.e. modifications can be done by updating the binaries and configuration on hosts where and when it’s needed. This mutable nature tempts engineers to make a temporary fix on any urgent issues. These fixes are often forgotten, resulting in lingering drift in configuration.
So, in order to avoid these issues hyperforce is built based on immutable infrastructure. In this infrastructure, the fix to the issue and its deployments are rolled out in an immutable way with the help of Software-driven Virtualized infrastructure, where every part of the infrastructure, network, and storage can be provided and managed dynamically via API calls.
So with the combined help of Virtual monitors and containers, the fix that needs to be made to the hosts are not made ‘in place’ instead the entire VM or Container is built and deployed as a newer version. - Infrastructure as Code: Software-driven Virtualized infrastructure – It is the combination of hardware and software, whereby hardware’s capacity can be managed based on the demand with the help of the software. This software will be using explicit metadata that is kept under source control.
So, when there is a change that needs to be made to the host, then we will have data regarding the updates made and It should follow the same protocol as any other part of a software update – validation, peer review, automated testing, staging, and gradual rollout. In this way we can reduce the bugs made by human error - Multi-Availability-Zone Design: To provide multiple availability, Salesforce uses numerous availability zones (AZs) on the public cloud. The commute resources like services and data storage technologies are stored in at least three different zones for each region with a separate infrastructure.
So if there is failure in any one of these zones due to some inevitable reasons the service can be provided with the help of other zones without interruption. - Zero Trust: In Hyperforce, Salesforce has implemented standard best security practices, which are automatically and consistently applied. The zero trust architecture means that no access to a resource is given until or unless it is authenticated, even for other components of its very own system.
On top of this hyperforce employs the principle of least privilege i.e. each operator is given just the right level of privilege and automated removal of that access after a period of time for production data.
Conclusion:
It is a win-win situation for both Salesforce and its customers because Salesforce does not need to maintain and upgrade its physical resources for data storage, and customers can benefit in many aspects like data residency, scalability, security, privacy, and agility
Resources:
Introducing Hyperforce – General Information & FAQ
Behind the Scenes of Hyperforce
How Salesforce Uses Immutable Infrastructure in Hyperforce
