Menu

Composable Infrastructure: Dynamically Changing IT Infrastructure

May 01, 2024

Composable Infrastructure:  Dynamically Changing IT Infrastructure

By Jaan Mannik, Director of Commercial Sales

The evolution of IT infrastructure spans several decades and is marked by significant advancements in computing technology, networking, storage, and management practices. Data Centers have historically relied on Converged or Hyper-Converged infrastructures when deploying their hardware, which proved limited in flexibility, efficiency, scalability, and support for the Artificial Intelligence / Machine Learning (AI/ML) modern workloads of today. Composable Disaggregated Infrastructure addresses these limitations by increasing resource utilization, and represents a paradigm shift in how IT infrastructure is designed. In this post, we’ll take an in-depth look at how this new technology is being deployed today, and why its revolutionizing the way organizations build and operate their IT environments.

In a Converged Infrastructure (CI), all the hardware resources are pre-configured and integrated into a single server, including processing, compute, storage, and networking tools. Multiple servers are racked together and connected over a common network backbone such as Ethernet, Fibre Channel, or InfiniBand. Converged Infrastructure is designed to simplify deployment by minimizing compatibility issues between systems, reduce complexity, and improve performance. For companies looking to scale, they simply deploy more pre-configured bundles of hardware and connect them to their network. While easier to manage, it often leads to scalability challenges, resulting in over-provisioning or underutilization of resources. Flexibility is also limited since deployments are often optimized for specific workloads, making it challenging to support changing requirements and introduce new technologies or components.

Composable Infrastructure

Hyper-Converged Infrastructure (HCI) addresses some of the limitations of a Converged Infrastructure by taking a software-based approach that integrates the processing, compute, storage, and networking resources into a single, software-defined, virtualized environment. Virtual servers are managed by a single hypervisor which provides simplified management, better scalability, and flexibility in modern data centers. Although a more adaptable IT Infrastructure approach, virtual machines (VMs) introduce performance overhead, latency, scaling granularity of a particular resource, and data locality concerns.

Enter Composable Disaggregated Infrastructure (CDI); a modern approach to IT Infrastructure that has transformed data center operations by providing a more agile, efficient, and flexible infrastructure foundation that can better support the evolving needs of modern businesses. Processing, compute, storage, and networking resources are disaggregated from their physical locations and put into a resource pool, which is managed by an API and dynamically allocated to a given workload while the other resources in the pool are being used elsewhere. For example, some workloads may require more GPUs for compute-heavy jobs or more storage for data recording purposes and those decisions can be made on the fly with software-defined intelligence which allows for optimal performance. CDI eliminates the need for workload-specific environments by offering a fluid set of resources that can be composed to meet the unique needs of any application. Resource utilization becomes optimized based on application requirements. Scalability becomes dynamic by being able to pick and choose resources in response to fluctuating workloads. Management is simplified since you don’t need to physically provision hardware. Downtime is also reduced because you can perform maintenance on unused resources without disrupting running applications.

Given all the advantages of composable infrastructure, many technology companies are competing to find an edge. Key differentiators between these technology companies often include bare metal servers, GPU/FGPA expansion accelerators, flash storage arrays, and networking interfaces, which are all interconnected by a physical top-of-rack PCI Express (PCIe) switch fabric and managed by their own special blend of software. PCIe switch fabrics are extremely high performance, supporting 256GB/s sustained throughput at Gen5, and provide much lower latency than other protocols such as Ethernet or InfiniBand. CDI companies take advantage of PCIe by leveraging expansion systems to support their GPU/FGPA/SSD/NIC resources and directly connect them to the top-of-rack PCIe switch. In doing so, all PCIe devices in the remote chassis appear to the host as directly connected PCIe peripherals on the PCIe switch fabric. A great example of the latest generation of PCIe 5.0 expansion system being leveraged by some of these technology companies today is the One Stop Systems' 4U Professional compute accelerator, which combines the power of the latest PCIe Gen5 add-in cards with an optimized, feature-rich, design for the most demanding HPC applications.

Overall, composable infrastructure represents a significant evolution in data center architecture, offering greater flexibility, efficiency, and scalability compared to traditional monolithic infrastructure models. By disaggregating resources into modular building blocks and providing a software-defined control plane for resource management, CDI enables organizations to adapt more quickly to changing business needs and drive innovation in their IT infrastructure.

Composable Infrastructure

Click the buttons below to share this blog post!

Return to the main Blog page




Leave a comment

Comments will be approved before showing up.


Also in One Stop Systems Blog

Edge Computing
The Four Types of Edge Computing

April 17, 2024

“Edge Computing” is a term which has been widely adopted by the tech sector. Dominant leaders in accelerated computing have designated “Edge” as one of their fastest-growing segments, with FY24 revenue projected to be nearly $100 billion. The boom in the market for Edge Computing has become so significant that it is increasingly common to see companies create their own edge-related spinoff terms such as ‘Rugged Edge’, ‘Edge AI’, ‘Extreme Edge’, and a whole slew of other new buzzwords. 

Continue Reading

Datalogging in Autonomous Military
Unveiling the Strategic Edge: Datalogging in Autonomous Military Vehicles

March 11, 2024

The landscape of modern warfare is undergoing a profound transformation with the integration of cutting-edge technologies, and at the forefront of this evolution are autonomous military vehicles. Datalogging, a seemingly inconspicuous yet indispensable technology, plays a pivotal role in shaping the capabilities and effectiveness of these autonomous marvels. In this blog post, we delve into the critical role of datalogging in autonomous military vehicles and its impact on the future of defense strategies.

Continue Reading

Redundancy and Management of Rugged Edge Servers
Redundancy and Management of Rugged Edge Servers

February 13, 2024 2 Comments

Computer server redundancy, including backup power supplies, RAID storage devices and applications that automatically fail-over, keeps critical systems up and running longer than non-redundant systems. Similarly, effective system monitoring can provide early warning of failures and allow system managers to remotely manage these systems, further improving application uptime. While the concepts of computer system redundancy and system management are well-established in all levels of computing, from the personal computer to the largest hyperscale datacenters, the unique challenges of placing datacenter-class computing elements performing AI applications in mobile edge environments, like aircraft, ships, and land vehicles, brings unique challenges to system redundancy and management. 

Continue Reading

You are now leaving the OSS website