Less complex and cloudy server fleet management with Nebulon – Blocks and Files

Nebulon is attracting customers with its Server Fleet Management Server Control Cards and Cloud Management, as customers realize they can save a lot of money by purchasing card-integrated servers.

The company briefed IT Press Tour attendees on its vision and progress in partner, customer and product development. Its technology is based on intelligent Storage Processing Units (SPUs), each connected to a host server via a PCIe 3 link and providing a way to boot the server, its direct attached storage, and manage its software and firmware environment. .

The briefing took place under Covid-19 restrictions and here a masked Craig Nunes holds a Nebulon swag t-shirt.

Co-founder and CEO Siamak Nazari said Nebulon partners with several server vendors — Dell, HPE, Lenovo, and Supermicro — and customers purchase servers as well as Nebulon cards from their strategic server vendor.

We asked about Cisco and COO Craig Nunes replied, “We haven’t spoken to them recently. He hinted that Cisco would rather sell its HyperFlex hyperconverged system than a UCS server + Nebulon card combination.

Nebulon aspires to partner with global systems integrators, such as Capgemini, but has no contracted resellers – they are contracted to server vendors and Nebulon’s invite-only smartPartner program aims to help them sell servers equipped with Nebulon cards with incentive discounts, training and demand creation programs, etc.

Nebulon states that it provides the Nebulon Intelligent Infrastructure for cloud repatriation and four products:

  • smartCore for VMware;
  • smartCore for Kubernetes;
  • smartiaaS for CSPs;
  • smartEdge – Single API, centralized fleet automation and monitoring [2-node HA starting footprint – zero-config cloud quorum].

He sees the appeal of large customers buying lots of servers that are loaded with server fleet management tasks, requiring servers to run them and admin time to update operating systems and firmware.

Nunes talked about a global CSP customer who had 250 VMware servers and 60 storage arrays of 2PB capacity, spread across 10 data centers. This CSP considered using HCI (hyperconverged infrastructure), but needed some of the HCI servers to run the infrastructure and thought the overhead was inefficient.

He decided to migrate to a Nebulon equipped infrastructure and the migration is about a third of the way there. The CSP has approximately 100 Nebulon-enabled servers and 1PB+ storage in four data centers in Europe and North America. Storage vMotion was used to move data from the arrays to the Nebulon-equipped servers, which have between 12 and 30 TB of capacity.

Upcoming features

Nazari said, “We allow 30TB of attached capacity per CPU. It will soon double to 60TB. He also said that Nebulon now supports PCIe 3 and will add PCIe 4 support in 2023.

Tobias Flitsch, Chief Product Officer, said that over the next two quarters, Nebulon will add:

  • Server management integration – centralized inventory and management of hardware and firmware across heterogeneous server vendors;
  • Nebulon Ansible Collection – tools to configure servers after deployment only with Ansible without the need for additional software / services (Ansible connects to the server loaded by Nebulon and loads the software elements of the system);
  • Instant cluster recovery – instantly recover an entire cluster.

The latter may be necessary when network monitoring of I/O patterns indicates a malware attack – ie. deduplication rates change or the I/O pattern changes from random to sequential. In the future, Nebulon may add machine learning to detect this and tell the client that it has detected abnormal behavior on the storage layer.