NVMe over TCP: How it supercharges SSD storage using standard IP networks

Soon after data centers began transitioning from hard drives to solid-state drives (SSD), the NVMe protocol arrived to support high-performance, direct-attached PCIe SSDs. NVMe was followed by NVMe over Fabrics (NVMe-oF), which was designed to efficiently support hyperscale remote SSD pools, effectively replacing direct-attached storage (DAS) to become the default protocol for disaggregated storage within a cloud infrastructure.To read this article in full, please click here

VMware, Nvidia offer GPU-powered AI in virtual machines

VMware and Nvidia have expanded their alliance to support Nvidia GPU-based applications on VMware's new vSphere 7 Update 2. The upgraded version of vSphere 7 will support the new Nvidia AI Enterprise offering, a suite of enterprise-grade AI tools and frameworks that enables GPU-accelerated applications to run in VMware virtual machines and containers.VMware's vSphere 7 U2 adds support for Nvidia's A100 Tensor Core GPU and its multi-instance GPU feature, which allows for partitioning of the cores on an A100 for use by multiple users, much in the same way VMware partitions CPU cores out to multiple users. Read more: Highflying Nvidia widens its reach into enterprise data centers To read this article in full, please click here

NVMe over TCP: How it supercharges SSD storage using standard IP networks

Soon after data centers began transitioning from hard drives to solid-state drives (SSD), the NVMe protocol arrived to support high-performance, direct-attached PCIe SSDs. NVMe was followed by NVMe over Fabrics (NVMe-oF), which was designed to efficiently support hyperscale remote SSD pools, effectively replacing direct-attached storage (DAS) to become the default protocol for disaggregated storage within a cloud infrastructure.To read this article in full, please click here

VMware, Nvidia offer GPU-powered AI in virtual machines

VMware and Nvidia have expanded their alliance to support Nvidia GPU-based applications on VMware's new vSphere 7 Update 2. The upgraded version of vSphere 7 will support the new Nvidia AI Enterprise offering, a suite of enterprise-grade AI tools and frameworks that enables GPU-accelerated applications to run in VMware virtual machines and containers.VMware's vSphere 7 U2 adds support for Nvidia's A100 Tensor Core GPU and its multi-instance GPU feature, which allows for partitioning of the cores on an A100 for use by multiple users, much in the same way VMware partitions CPU cores out to multiple users. Read more: Highflying Nvidia widens its reach into enterprise data centers To read this article in full, please click here

SD-WAN may be the key to smart network services

If you stop and think, a lot of our expectations about network services are really about personality—our own.  We’d like our services to work, well, the way we work.  We’d like them to know us, to tune to our needs, right?  Do you think that some giant global interconnect with hundreds of thousands of elements is going to be able to do that?  Nope, which means personalized services will have to come down to the only piece we really own—the lowly network edge.We learned decades ago that you can’t make giant networks user- or service-aware.  Awareness of this sort, which is known as “statefulness” in network-speak, means sticking little pieces of a virtual-you into the network to represent your interests. Maybe these pieces are an entry in a routing table, or maybe they’re a policy stored in some repository and sent to the devices that handle your traffic, but they’re individualized if what they’re doing is to personalize.  That just doesn’t scale.  Not only are there too many little pieces, network traffic could get reconfigured or a device could fail, and all at once your personalizing pieces aren’t even where your traffic Continue reading

Worth Reading: Modules, Monoliths, and Microservices

If you want to grow beyond being a CLI (or Python) jockey, it’s worth trying to understand things work… not only how frames get from one end of the world to another, but also how applications work, and why they’re structured they way they are.

Daniel Dib recently pointed out another must-read article in this category: Modules, monoliths, and microservices by Avery Pennarun – a wonderful addition to my distributed systems resources.

Worth Reading: Modules, Monoliths, and Microservices

If you want to grow beyond being a CLI (or Python) jockey, it’s worth trying to understand things work… not only how frames get from one end of the world to another, but also how applications work, and why they’re structured they way they are.

Daniel Dib recently pointed out another must-read article in this category: Modules, monoliths, and microservices by Avery Pennarun – a wonderful addition to my distributed systems resources.

pygnmi 8. Securing the gNMI connectivity with self-signed certificates.

Hello my friend,

Continuing our explanation of the pyGNMI, we’ll take a loon into the security aspect of the tool. Namely, we will take a look how quickly and easily you can implement the encryption between your host running pyGNMI and the gNMI speaking network function.


1
2
3
4
5
No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.

Where is the GNMI in the Network Automation?

The automation world (network and not only) can at a high level be split into text-based automation and model-driven automation. The text based automation is all about Linux systems, where we typically template the whole configuration files, put them in the corresponding folders and then restart daemons. The model driven approach is where we communicate with the network devices using the NETCONF, RESTCONF or gNMI based on the YANG modules. At the current moment, gNMI is most dynamically developing protocol. We not only teaches you how it works and when to use it, but we also created a Python library, Continue reading

Who Pays The Price of Redundancy?

No doubt by now you’ve seen the big fire that took out a portion of the OVHcloud data center earlier this week. These kinds of things are difficult to deal with on a good day. This is why data centers have reductant power feeds, fire suppression systems, and the ability to get back up to full capacity. Modern data centers are getting very good at ensuring they can stay up through most events that could impact an on-premises private data center.

One of the issues I saw that was ancillary to the OVHcloud outage was the small group of people that were frustrated that their systems went down when the fire knocked out the racks where their instances lived. More than a couple of comments mentioned that clouds should not go down like this or asked about credit for time spent being offline or some form of complaints about unavailability. By and large, most of those complaining were running non-critical systems or were using the cheapest possible instances for their hosts.

Aside from the myopia that “cloud shouldn’t go down”, how do we deal with this idea that cloud redundancy doesn’t always translate to single instance availability? I think we Continue reading

Heavy Networking 566: Inside Intel’s Strategy To Unlock Data Center Performance (Sponsored)

On today's Heavy Networking we dive into Intel's portfolio---including Tofino, SmartNICs, P4, and more---to understand how it unlocks the compute power of your data center. Our guest is Mike Zeile, Data Center Group Vice President and General Manager of End-to-End Network Applications at Intel. Intel is our sponsor.

The post Heavy Networking 566: Inside Intel’s Strategy To Unlock Data Center Performance (Sponsored) appeared first on Packet Pushers.

Technology Short Take 138

Welcome to Technology Short Take #138. I have what I hope is an interesting and useful set of links to share with everyone this time around. I didn’t do so well on storage links; apologies to my storage-focused friends! However, there should be something for most everyone else. Enjoy!

Networking

  • I’ve been interested in learning more about gRPC, so this guide on analyzing gRPC messages using Wireshark may be useful.
  • Isovalent, the folks behind Cilium, recently unveiled the Network Policy Editor, a graphical way of editing Kubernetes Network Policies.
  • Ivan Pepelnjak, the font of all networking knowledge, has been discussing cloud networking in some detail for a good while now. The latest series of posts (found here and here) are, in my opinion, just outstanding. I want to be like Ivan when I grow up. #BeLikeIvan
  • If you work with TextFSM templates (see here for more information), then you might also like this post on writing a vim syntax plugin for TextFSM templates.
  • Want/need to better understand IPv6? Denise Fishburne has you covered. Denise also has you covered if you need BGP knowledge.

Security

In Macedonia, Strengthening IXP.mk’s Peering Infrastructure

The Internet Society has been supporting the development of the Internet in Macedonia by collaborating with the Faculty of Computer Science and Engineering (FCSE) of the Saints Cyril and Methodius University in Macedonia on its IXP.mk project. IXPs play a critical role in bringing faster and more affordable Internet, and the Macedonian IXP (IXP.mk) had been established in June 2018 with technical support from a number of stakeholders.

Switch and MUX to Strengthen Infrastructure

As traffic grew, IXP.mk needed to increase its peering capacity with improved switching capabilities and space for data racks that would allow it to attract new participants to the exchange. In 2020, the Internet Society provided IXP.mk with a switch and two Fiber Optic Multiplexers (Fiber Mux) that enabled an additional peering location to be established in the Telesmart Telekom data center, thereby making it easier for other major networks and Content Distribution Networks to peer with each other. The Fiber Muxes support the transmission of multiple data channels over a single fiber that has been donated by an existing member of IXP.mk.

Critical Service Provider

With a strong peering infrastructure, IXP.mk is now a critical service provider to Continue reading

Lessons Learned from Scaling Up Cloudflare’s Anomaly Detection Platform

Introduction to Anomaly Detection for Bot Management

Lessons Learned from Scaling Up Cloudflare’s Anomaly Detection Platform

Cloudflare’s Bot Management platform follows a “defense in depth” model. Although each layer of Bot Management has its own strengths and weaknesses, the combination of many different detection systems — including Machine Learning, rule-based heuristics, JavaScript challenges, and more — makes for a robust platform in which different detection systems compensate for each other’s weaknesses.

One of these systems is Anomaly Detection, a platform motivated by a simple idea: because bots are made to accomplish specific goals, such as credential stuffing or content scraping, they interact with websites in distinct and difficult-to-disguise ways. Over time, the actions of a bot are likely to differ from those of a real user. Anomaly detection aims to model the characteristics of legitimate user traffic as a healthy baseline. Then, when automated bot traffic is set against this baseline, the bots appear as outlying anomalies that can be targeted for mitigation.

An anomaly detection approach is:

  • Resilient against bots that try to circumvent protections by spoofing request metadata (e.g., user agents)
  • Able to catch previously unseen bots without being explicitly trained against them.

So, how well does this work?

Today, Anomaly Detection processes more than Continue reading

Cisco bumps up its silicon speed to 25.6Tbps

Cisco continues to crank up the speed of its webscaler-class Silicon One devices, this week adding three new devices—including 25.6Tbps switching silicon it says delivers  1.7 times higher bandwidth and three times higher packets-per-second than other silicon on the market.Cisco unveiled the Silicon One family of unified switches and routers in 2019 as part of what it called its “Internet for the Future”  strategy.  The Silicon One family is important as Cisco has designs on being a leading provider of the network underpinnings of large webscale and service provider networks. The family can also help Cisco compete effectively with others such as Intel, Broadcom, Juniper, Arista.To read this article in full, please click here