Skip to main content
  • Place orders quickly and easily
  • View orders and track your shipping status
  • Enjoy members-only rewards and discounts
  • Create and access a list of your products
  • Manage your Dell EMC sites, products, and product-level contacts using Company Administration.

Article Number: 000146848

Extreme Scale Infrastructure (ESI) Architects' POV Series: Redfish

Summary: This article provides information about Redfish Infrastructure

Article Content


Authored by: Paul Vancil – Distinguished Eng., ESI Architecture, Dell EMC, 8 Sep 2017, 8:11

In today’s typical large scale data center the numbers of servers that organizations are trying to keep track of have reached enormous proportions. In addition to that, existing protocols lack the ability to describe modern architectures and do not incorporate modern security best practices. The differences between various vendors’ solutions only multiplies the problem. Organizations have been forced to manage to the lowest common denominator – severely compromising their capabilities.

Redfish origins and Dell EMC

To address these challenges, in 2014, the Distributed Management Task Force created its Scalable Platform Management Forum (SPMF) to begin work on Redfish, an open, industry standard specification and schema to provide a consistent API that addresses all the components in the data center – one that is simpler, more secure and uses modernized protocols. Redfish is an ongoing project that is supported by participation from a wide array of server vendors, including Dell EMC, Microsoft, Intel, HPE, Lenovo and IBM. Dell has co-chaired the SPMF since its inception, myself from 2014 until late last year, and now my Dell EMC colleague, Michael Rainieri is co-chair.

Version 1.0

The initial goal for Redfish 1.0 was providing an IPMI 2.0 equivalence with extensions for processor, NIC, and storage inventory - essentially standardizing the level of functionality available in IPMI – but with a benefits of a more secure RESTful interface. . Much of the early work went into establishing the protocol and modeling rules. Considering the complexity of the task, and the number of industry participants, the project held to an extremely aggressive schedule and delivered Redfish v1.0 in August of 2015, just a year after the forum began its work.


Redfish impact on the industry

Not only was the initial release held to a tight timeline – but since then, and key to the success of Redfish, there has been wide spread acceptance and adoption of the APIs by industry vendors. Most of the major server vendors are currently implementing some level of the Redfish APIs. Dell EMC has been aggressive in its implementation of Redfish – PowerEdge servers (including 2 generations back) have support for v1.0 and parts of subsequent releases, and the Rack Manager component on the DSS 9000 uses Redfish as the basis of its rack-wide management capabilities.

Having a de facto standard for system management APIs is a boon for data center management. One of the most significant benefits is that system management software and orchestration tools now have a standard way to address heterogeneous systems in large scale infrastructures. This both simplifies the development effort and expands the capabilities of administrators. Because open standards development involves wide participation from across the industry, adopters of the standard benefit from the industry-wide expertise and input and more work can be done more quickly.

Redfish and composability

The fact that customers across the industry were having to wrestle with the mentioned management challenges helped drive the project to this point, but so did the advent of Intel® Rack Scale Design (RSD). RSD is an architecture that includes a multi-rack hardware resource manager and has been developed in the same timeframe as Redfish. It uses Redfish APIs as its standard way of accessing hardware resource information. Using Redfish APIs and RSD concepts, data centers can dynamically allocate pooled hardware resources. These two elements working together are foundational steps on the path to fully composable infrastructure.

The Redfish roadmap

Redfish v1.0 proved that it was possible to finally get the development community to standardize on basic management APIs. But the ultimate goal of the Redfish project is to provide even higher level capabilities to the modern data center. To that end, the SPMF has committed to delivering three release bundles each year with additional capabilities. Work is in-progress to define standards for: Ethernet Switches, Data Center Infrastructure (e.g., PDUs, BBUs), External Storage (via SNIA), and Composability. In addition, the SPMF is continuing to update interoperability capabilities with conformance test tools, and interoperability feature profiles that define the minimum required features for particular environments; for example, a Base Server feature profile that would standardize the minimum set of guaranteed Redfish features for a base server Interoperability is really what Redfish is all about, and Dell EMC is actively working as part of this industry effort to help bring it to our data center customers.

For more details on all of the Redfish activities and releases see the Distributed Management Task Force website.

Article Properties

Last Published Date

10 Apr 2021



Article Type