Cloud computing offers a multitude of possibilities for system architecture, including highly available and redundant systems. However, the challenge of latency can impact the functionality of these systems. Many cloud service providers use availability zones (AZs) to create highly available architectures, but the latency between these zones can be a significant hurdle for workloads that require high availability and resiliency. IBM Cloud has introduced the Madrid multi-zone region (MZR), which reduces latency between AZs to less than 1ms, allowing highly available workloads to operate at on-premises speeds without sacrificing the benefits of geographic distribution.

Resiliency and availability in the cloud: A latency challenge

While the cloud offers near-infinite infrastructure, it is still limited by geographic distance and time. Resilient systems can be built on the cloud, but they often struggle to meet system requirements due to latency concerns. Availability zones, which are commonly used by cloud service providers, are split across a specific geographic region, typically tens of miles apart. This setup provides low single-digit latency between AZs in the same region. However, when it comes to high availability and resiliency, this latency can present significant challenges.

In the example of the fictional NeverNeverEver (NNE) service, which provides a redundant storage platform, requests traverse AZ boundaries multiple times, resulting in increased latency. By reducing latency between AZs, the total processing time for each request can be significantly decreased.

Reducing latency by 75%+

IBM Cloud’s Madrid MZR offers sub-1ms latency between availability zones, eliminating concerns about latency for highly available workloads. With these reduced latency rates, the request processing time can be reduced by more than 75%, resulting in faster transactions.

To achieve true on-premises performance, resilience, and cloud scale and security, enterprises can build their high availability workloads in IBM Cloud’s Madrid MZR.

Getting started with IBM Cloud VPC

New and existing IBM Cloud customers can easily sign up and start using IBM Cloud VPC. By using the code VPC1000, customers can receive up to USD 1,000 in credits for VPC infrastructure components, including networking, storage, and compute.

Sources:

• IBM Blog

FAQ

What is latency in cloud computing?

Latency refers to the delay or lag between the time a request is made and the time that the response is received. In cloud computing, latency can be affected by factors such as geographic distance, network congestion, and processing time.

Why is latency a concern for highly available and redundant systems?

Latency is a concern for highly available and redundant systems because it can impact the performance and functionality of these systems. If the latency between different components or availability zones is high, it can result in delays and decreased responsiveness, which can be detrimental to the availability and resilience of the system.

How does the Madrid multi-zone region (MZR) reduce latency?

The Madrid MZR offered by IBM Cloud reduces latency between availability zones to less than 1ms. This allows highly available workloads to operate at on-premises speeds, enabling faster transactions and improved performance.

What are availability zones?

Availability zones are separate data centers within a geographic region that are designed to provide redundancy and high availability. They are typically located in different physical locations and connected by high-speed, low-latency networks. Availability zones allow for the distribution of workloads and data across multiple data centers, ensuring that if one zone goes down, the workload or data can be seamlessly transferred to another zone.

What are the benefits of using IBM Cloud VPC?

IBM Cloud VPC offers a secure and dedicated virtual private cloud environment that enables enterprises to deploy and manage their applications and services with enhanced control and flexibility. It provides a scalable and highly available infrastructure for running workloads, along with features such as network isolation, high-performance storage, and built-in security.