Greg Ferro

Author Archives: Greg Ferro

Research: Wired Ethernet: Intel® Ethernet X520 to XL710 -… |Intel Communities

This balance is also important when looking at the interaction within a server between the network cards (which have some on-board buffering) and the DPDK managed buffer resources on the host. A better tuning of the buffer sizes can eliminate potential packet losses. This paper is summarizing what to do when going from one type of network card to another one that has different on-board buffer behavior. It also has the potential to explain and fix certain packet loss issues going from one generation of a NIC card to another (e.g. when moving from Intel® Ethernet Server Adapter X520 to Intel® Ethernet Controller XL710)

Basically it comes down to configuring the RX descriptors.

So, to avoid packet losses due to CPU core being interrupted when using Fortville (or when using Niantic and SRIOV), the number of RX descriptors should be configured high enough, for instance to 2048.

Wired Ethernet: Intel® Ethernet X520 to XL710 -… |Intel Communities : https://communities.intel.com/community/wired/blog/2017/01/09/intel-ethernet-x520-to-xl710-tuning-the-buffers-a-practical-guide-to-reduce-or-avoid-packet-loss-in-dpdk-applications

Link to local version PDF File for my future self (hi there!)

X520_to_XL710_Tuning_The_Buffers.pdf

The post Research: Wired Ethernet: Intel® Ethernet X520 to XL710 -… |Intel Communities appeared first on EtherealMind.

BT Openreach to trial Dark Fibre Access in August 2017

When the government owns the fibre and copper cabling but leaves the operation and revenue extraction to commercial companies you get competitive telecoms

Openreach’s DFA product will allow rival internet service providers (ISPs) access to the company’s fibre-optic cables, which technically means that they can install their own equipment at either end of the optical fibre, within Openreach’s cable ducts.

BT Openreach to trial Dark Fibre Access in August 2017 : https://www.telegeography.com/products/commsupdate/articles/2017/01/24/bt-openreach-to-trial-dark-fibre-access-in-august-2017/

The post BT Openreach to trial Dark Fibre Access in August 2017 appeared first on EtherealMind.

Avaya Inc. Files for Chapter 11 Protection

Not much of a surprise, the Avaya business has been doing OK but the company has a large about of debt that is dragging it down. Chapter 11 likely to allow restructuring of debt and plan a path forward.

Its possible that the business could be broken up or pieces sold off to pay down debt but the lack of buyers for Brocade (either whole or in part) suggests that is unlikely for networking. Other parts of Avaya might be sold off

Link: Avaya Inc. Files for Chapter 11 Protection – http://www.avaya.com/en/about-avaya/newsroom/news-releases/2017/pr-us-170119a/

Customer FAQ: http://www.avaya.com/en/documents/filing-faqs.pdf has more info:

As a result of the terms of Avaya’s debt obligations and the upcoming debt maturities, we need to recapitalize the Company and believe the restructuring process is the best path forward at this time. Our businesses are healthy and performing well, and we are executing at a high level.

Seems likely.

The post Avaya Inc. Files for Chapter 11 Protection appeared first on EtherealMind.

Dictionary: Despondent

Despondent: a deep dejection arising from the conviction of the uselessness of further effort.

A perfect word of current state of Enterprise IT.

Definition of despondent : feeling or showing extreme discouragement, dejection, or depression

despondent : in low spirits from loss of hope or courage

The post Dictionary: Despondent appeared first on EtherealMind.

Musing: ITC rejects de facto standard defense (337-TA-944, Cisco v. Arista) | Essential Patent Blog

Detailed but accessible legal review of Cisco vs Arista case. Dated July 2016 so it doesn’t cover the latest developments but provides a lot of insight into the legal.

My current view on this issue:

  1. Cisco is validating Arista as a serious, viable competitor. On balance, Arista gets more out of this than Cisco does and doing a nice job of being the under dog.
  2. Rumours suggest that its personal matter to attack Arsita for some executives and not a business matter.
  3. Customers perceive Cisco as wasting time and energy on legal matters instead of innovating new products or improving product quality.
  4. Customers money is being wasted on lawyers instead of solutions.

In December 2015, Cisco Systems, Inc. (Cisco) filed a complaint alleging that certain network devices (switches) imported by Arista Networks, Inc. (Arista) infringed several Cisco patents directed to computer networks.  Arista raised several equitable defenses based, in part, on allegations that Cisco submitted a request for comments document RFC 5517 to IETF and promoted RFC 5517 to the public generally as an “informal standard” for private virtual local area networks (PVLANs) for which Cisco would not assert its patents or would license on fair, reasonable and non-discriminatory (FRAND) Continue reading

Response: Saving you bandwidth on Google+ through machine learning

Image compression reduces network requirements in a major way.

  1. Smaller files means less bandwidth
  2. Less round trips for faster page rendering
  3. reduces load on network function like proxies, IDS, etc.

And its already in production for Android devices using Google services.

To help everyone be able to see the beautiful photos that photographers share to Google+ in their full glory, we’ve turned to machine learning and a new technology called RAISR. RAISR, which was introduced in November, uses machine learning to produce great quality versions of low-resolution images, allowing you to see beautiful photos as the photographers intended them to be seen. By using RAISR to display some of the large images on Google+, we’ve been able to use up to 75 percent less bandwidth per image we’ve applied it to.

Saving you bandwidth on Google+ through machine learning

The post Response: Saving you bandwidth on Google+ through machine learning appeared first on EtherealMind.

Arris in bid for Brocade network unit: sources | Reuters

Arris wants to buy the Ruckus wireless business.

Arris is looking to buy Brocade’s network edge business, which is the most valuable of the assets being sold, according to the sources. Arris is not in talks to buy other parts of the business being divested by Brocade that include data centers, switching and software, the sources added.

Apparently talks to buy other parts of the business in whole or part are not working out.

Broadcom had divided up Brocade’s divestiture into three pieces after an earlier deal to sell the whole business to a private equity firm fell apart late last year, according to the sources. A private equity firm could still step up to buy all three pieces, the people said.

Arris in bid for Brocade network unit: sources | Reuters

The post Arris in bid for Brocade network unit: sources | Reuters appeared first on EtherealMind.

Response: Codec 2 700C | Rowetel

Acceptable quality open source voice codec in 700 bps

My endeavor to produce a digital voice mode that competes with SSB continues. For a big chunk of 2016 I took a break from this work as I was gainfully employed on a commercial HF modem project. However since December I have once again been working on a 700 bit/s codec. The goal is voice quality roughly the same as the current 1300 bit/s mode. This can then be mated with the coherent PSK modem, and possibly the 4FSK modem for trials over HF channels.

Codec 2 700C | Rowetel

The post Response: Codec 2 700C | Rowetel appeared first on EtherealMind.

Response: Introducing Open/R — a new modular routing platform | Engineering Blog | Facebook Code | Facebook

Although this post is from May 2016, Petr Lapukhov at Facebook outlines an method to replace routing protocols with a message bus to enable real network applications.

I’m doubtful that wider networking market would adopt something that doesn’t have BGP in the solution but Facebook has the resources to develop something like this and prove that it works. That could change perceptions. In any case, thought provoking reading.

Introducing Open/R — a new modular routing platform | Engineering Blog | Facebook Code | Facebook: “The Open/R software enables rapid prototyping and deployment of new applications to the network much more frequently than the industry’s standard development process. To create an interoperable standard, the industry’s process is often lengthy due to code being built independently by multiple vendors and then slowly deployed to their customer networks. Furthermore, every vendor has to accommodate for the demands of numerous customers — complicating the development process and requiring features that are not always useful universally.”

The post Response: Introducing Open/R — a new modular routing platform | Engineering Blog | Facebook Code | Facebook appeared first on EtherealMind.

Response: Coming soon with Cumulus Linux 3.2: EVPN

Increasingly coming to the view that BGP-EVPN is a big deal. Neither vendors or customers can imagine their networks without a 30 year old routing protocol so this is the half-pregnant, half-arsed solution that seems likely to gain widespread adoption.

You can mangle BGP configuration with an application and call it SDN. Heck, IXPs have been doing that for a decade so its not new.

Welcome to networking where “its not new” is the byline for SDN.

Coming soon with Cumulus Linux 3.2: EVPN – Cumulus Networks Blog: “Can you summarize the benefits of deploying EVPN?

Cumulus EVPN provides many benefits to a data center, including:

Controller-less VXLAN: No controller is needed with EVPN, as it enables VTEP peer discovery through BGP.
Scale and Robustness: EVPN uses the standard BGP routing protocol for the control plane. BGP is a mature well-known protocol that powers the internet. For data centers that already run BGP, this involves just adding another address-family.
Fast convergence/mobility: The BGP EVPN address family includes features to track host moves across the datacenter, allowing for very fast convergence.
Multi-vendor interoperable: Since EVPN is a standard, it will be interoperable with other vendors that adhere to the Continue reading

Response: Japan researchers warn of fingerprint theft from ‘peace’ sign

Another one from the Biometrics is not useful for authentication dumpster:

The NII researchers were able to copy fingerprints based on photos taken by a digital camera three metres (nine feet) away from the subject.

Japan researchers warn of fingerprint theft from ‘peace’ sign : http://phys.org/news/2017-01-japan-fingerprint-theft-peace.html

The post Response: Japan researchers warn of fingerprint theft from ‘peace’ sign appeared first on EtherealMind.

Response: Engineers know how to party! Start your own celebration with Mellanox Ethernet solutions – YouTube

I approve of this positive, supportive and realistic representation of network engineers.

Engineers know how to party! Start your own celebration with Mellanox Ethernet solutions – YouTube: “”

The post Response: Engineers know how to party! Start your own celebration with Mellanox Ethernet solutions – YouTube appeared first on EtherealMind.

Dictionary: optic boom

optic boom

A flash produced when electrons move faster than light, akin to the boom of supersonic jets. Breaking the “light barrier” sounds like sci-fi, but physicists say it can happen in graphene sheets. The discovery could spark development of optical circuits a million times faster than silicon chips.

Link: The 21 Best New Words of 2016 | WIRED https://www.wired.com/2016j/12/21-best-new-words-2016/

The post Dictionary: optic boom appeared first on EtherealMind.

Response: Proposed server purchase for GitLab.com | GitLab

Gitlab is talking about heading into the private cloud after successfully building a cloud-ready application. The savings are substantial for a small, technology-rich company:

The cloud hosting for GitLab.com excluding GitLab CI is currently costing us about $200k per month. The capital needed for going to metal would be less than we pay for 1 quarter of hosting. The hosting facility costs look to be less than $10k per month. If you spread the capital costs over 2.5 years (10 quarters) it is 10x cheaper to host your own. (My emphasis)

This sounds about right but I don’t think this factors in head count for operating the physical infrastructure. Lets say that two extra FTEs at $15K per month are required, this still one third the cost of AWS. The reaility is $2.4MM is a substantial yearly budget for IT Infrastructure and for an application that already cloud-ready it would go a very long way

For a small company that is focussed on technology adding more headcount is good for capacity. In a team of ten people, adding 2 headcount increases diversity of thinking, ideas and approaches and can be important to spreading out the workload e. Continue reading

Response: The Orphaned Internet – Taking Over 120K Domains via a DNS Vulnerability in AWS, Google Cloud, Rackspace and Digital Ocean | The Hacker Blog

So obvious but I’m checking my unused domains to make sure they have no nameservers configured

The root of this vulnerability occurs when a managed DNS provider allows someone to add a domain to their account without any verification of ownership of the domain name itself. This is actually an incredibly common flow and is used in cloud services such as AWS, Google Cloud, Rackspace and of course, Digital Ocean. The issue occurs when a domain name is used with one of these cloud services and the zone is later deleted without also changing the domain’s nameservers. This means that the domain is still fully set up for use in the cloud service but has no account with a zone file to control it. In many cloud providers this means that anyone can create a DNS zone for that domain and take full control over the domain. This allows an attacker to take full control over the domain to set up a website, issue SSL/TLS certificates, host email, etc. Worse yet, after combining the results from the various providers affected by this problem over 120,000 domains were vulnerable (likely many more).

The Orphaned Internet – Taking Over 120K Domains via Continue reading

Musing: Norton Core Router | Introducing the Future of WiFi.

We set out to change WiFi forever. Our uncompromising vision: a wireless router that secures your digital life, while delivering the highest level of performance. The result is Norton Core. Core uses advanced machine learning and Symantec’s global intelligence network to defend your home WiFi—and every device connected to it—against malware, viruses, hackers and much more.

Powerful, secure WiFi is now a thing.

  1. We know that Norton produces sub-standard, low quality, deeply insecure anti-virus & malware desktop software. How good do you think this is ?
  2. Want to bet that no one cares and the branding plus a lot of marketing dollars will make it successful
  3. Attempting to be a full security suite in a single box. Oh, and WiFi.
  4. Enterprise-grade security features can now be offered to retail customers – (Enterprise comes last)
  5. Uses cloud for intelligence and threat analysis to power the security engine. Reuse of existing technology means its cheap and profitable.
  6. MIMO is hard, beam forming is really hard. I’m doubtful that Norton could get this right the first time (or at all).
  7. Lots of effort has been spent to make it look nice. Which is nice.
  8. Its probably better than nothing (which is what you really Continue reading

Response: XenDesktop and XenApp Interoperability with VMware NSX

Citrix and NSX integration to build departmental isolation between VDI desktops.

In this blog, I am going to focus on XenDesktop, NetScaler and NSX interoperability. I will discuss a field use case, see how to implement that in VMware NSX for XenDesktop and then look at a few micro-segmentation deployment scenarios, to showcase how XenDesktop and NetScaler in conjunction with NSX provides a compelling deployment model.

The ability to isolate and control desktops represents a major security and integrity enhancement. Using a Netscaler provides a single point of access into the data centre and can be integrated into the NSX overlay.

Adding NetScaler in this deployment would simplify the set up and allow the users of all the airlines (or tenants) access the same landing URL and still have complete isolation from each other’s data and resources.

Observation: NSX is an automation tool for connectivity between end points and offers isolation/segmentation as service. Added to Netscaler, we get orchestration to produce greater business benefits because the XenDesktop & NSX becomes a unified service.

No specific provisioning to make this happen. Thats an SDN outcome.

XenDesktop and XenApp Interoperability with VMware NSX | Citrix Blogs

The post Response: XenDesktop and XenApp Continue reading

1 … 45 46 47 48 49 … 71