Non Converged vs. Converged vs. Hyperconverged Infrastructure
In this post I want to give you a summary about what all these terms (marketing buzzwords) in the title meant and what’s all behind, I also want to show the differences between them.
General definition of converged
converged -> approach each other
What is a Hyperconverged Infrastructures (HCI)?
A hyperconverged infrastructure (HCI) is an architectural design to combine pools of server and storage resources into a 100% software-defined solution (appliance or single SKU).
By this design the separate infrastructure components are fully decoupled from the hardware in order to administrate and monitor them centrally.
So your complete IT Infrastructure can be managed, configured, scaled, automatized and monitored by only one central user interface.
Hyper-converged infrastructure (HCI) is a software-defined IT infrastructure that virtualizes all of the elements of conventional “hardware-defined” systems. HCI includes, at a minimum, virtualized computing (a hypervisor), software-defined storage, and virtualized networking (software-defined networking). HCI typically runs on commercial off-the-shelf (COTS) servers.
The primary difference between converged infrastructure (CI) and hyper-converged infrastructure is, that in HCI both, the storage area network and the underlying storage abstractions are implemented virtually in software (at or via the hypervisor) rather than physically in hardware. Because all of the software-defined elements are implemented in the context of the hypervisor, management of all resources can be federated (shared) across all instances of a hyper-converged infrastructure.
What is a Converged Infrastructures (CI)?
A traditional multi-tiered datacenter (aka, legacy or non converged infrastructure) requires by design that server (compute), storage and networking components be individually configured and linked together. Usually a dedicated IT team would be responsible for managing a single component which can become costly and complex.
In contrast a converged infrastructure (CI) will deliver all of these separate components build and unified into one physical appliance out of the box. Therefore enterprises will reduce costs associated with cabling, cooling, power and hiring dedicated teams of IT professionals.
Converged infrastructure (CI) is a way of structuring an information technology (IT) system which groups multiple components into a single optimized computing package. Components of a converged infrastructure may include servers, data storage devices, networking equipment and software for IT infrastructure management, automation and orchestration.
IT organizations use converged infrastructure to centralize the management of IT resources, to consolidate systems, to increase resource-utilization rates, and to lower costs. Converged infrastructures foster these objectives by implementing pools of computers, storage and networking resources that can be shared by multiple applications and managed in a collective manner using policy-driven processes.
IT vendors and IT industry analysts use various terms to describe the concept of a converged infrastructure. These include “converged system”, “unified computing”, “fabric-based computing”, and “dynamic infrastructure”.
Non Converged vs. Converged vs. Hyperconverged Infrastructure
A traditional multi-tiered data center (aka, legacy or non converged infrastructure) requires by design that server (compute), storage and networking components be individually configured and linked together. Usually a dedicated IT team would be responsible for managing a single component which can become costly and complex.
This design/architecture is basically the same as a Converged Infrastructure (CI). There’s just one major difference between them, a Converged Infrastructure (CI) is usually one out-of-the-box resp. pre-assembled appliance ready to use by a vendor.
A converged infrastructure (CI) in contrast and as mentioned will deliver all of these separate components from the non converged infrastructure build and unified into one out-of-the-box resp. pre-assembled appliance (single SKU) ready to use by a vendor. Therefore enterprises will reduce costs associated with cabling, cooling, power and hiring dedicated teams of IT professionals.
On the other hand, hyperconverged infrastructure (HCI) uses intelligent software to combine x86-based server and storage resources into a 100% software-defined solution.
A converged product is nothing but some sort of appliance out-of-the-box.
Broken down to its simplest, each server will also provide the prerequisites to function as as converged product. Servers nowadays supports to include a bunch of direct attached SSDs (storage), multiple CPUs (compute) with a vast amount of RAM and a bunch of separate network cards (network), not to mention the virtual networks of a hypervisor.
Both designs, converged infrastructure (CI) and hyperconverged infrastructure (HCI), merge the server (compute), storage and networking components in a data center but will achieve this in a different way.
Further both designs will also simplify the management of your IT Infrastructure.
In a nutshell the main difference between both designs is that hyperconverged infrastructure (HCI) also incorporates the storage part into the server (compute). The servers itself will linked with software to clusters what in past was only achieved by using a SAN.
By this approach the hardware is consolidated and complexity will be reduced here. The main part is now focused to the software to manage the appliance.
What is a Workload?
A workload refers to the amount of processing that a computer system or network handles at any given time. It is a measure of the demand placed on the system’s resources, including CPU, memory, storage, and network bandwidth, by the applications and processes running on it.
Workloads can vary widely depending on the nature of the tasks being performed and the specific applications running on a system. Different workloads may have distinct characteristics, such as computational intensity, storage requirements, and network communication patterns.
A workload can be any application that runs on a computer. These days the terms workload, application, software and program are used interchangeably.
Here are a few examples of different types of workloads:
- Compute Workloads: These workloads involve tasks that require significant CPU processing power. Examples include scientific simulations, rendering, and complex calculations.
- Storage Workloads: Workloads that involve a high level of data read and write operations to storage devices. Databases, file servers, and applications with large data sets often generate storage-intensive workloads.
- Network Workloads: Workloads that depend heavily on network communication. This includes applications that transfer large amounts of data over the network, such as video streaming, file transfers, or communication between distributed components.
- Memory Workloads: Some applications and processes are particularly demanding on system memory (RAM). For instance, in-memory databases and applications that manipulate large data sets in memory generate memory-intensive workloads.
- Mixed Workloads: Many real-world applications involve a combination of compute, storage, network, and memory demands. Mixed workloads require a balanced approach to resource allocation.
In summary, a workload in IT refers to the amount and nature of processing demands placed on a computer system or network by the applications and processes it is running.
Converged vs Hyperconverged Infrastructure
Converged vs. Hyperconverged Infrastructure
Dell EMC Integrated System for Microsoft Azure Stack HCI
Hyper-converged infrastructure (HCI)
Dell CONVERGED INFRASTRUCTURE
Converged Infrastructure Webinars