Technology Short Take 140

Welcome to Technology Short Take #140! It’s hard to believe it’s already the start of May 2021—my how time flies! In this Technology Short Take, I’ve gathered some links for you covering topics like Azure and AWS networking, moving from macOS to Linux (and back again), and more. Let’s jump right into the content!

Networking

Servers/Hardware

Security

Knowledge is Powerful and Needs to Be Shared

WritingPen

A tweet this morning from my friend Stephanie stood out in my timeline because she’s talking about something I’ve seen happen over and over again in my lifetime:

How many times have we seen this in our organizations? People want to hoard knowledge because they feel like it’s power. Maybe they’re worried that if anyone knew what they know it would mean they could get fired. Perhaps they enjoy holding the keys to the kingdom and not allowing anyone else to do something or know something they know. It could even be that they like the idea of mystery in the air and not allowing people to know the whole truth keeps things alive, as the founders of Coca-Cola and Colonel Sanders will happily tell you.

Over the years I’ve figured out that hoarding knowledge leads to ruin. I’ve been involved in so many scenarios were a lack of knowledge sharing ended up Continue reading

The Hedge 82: Jared Smith and Route Poisoning

Intentionally poisoning BGP routes in the Default-Free Zone (DFZ) would always be a bad thing, right? Actually, this is a fairly common method to steer traffic flows away from and through specific autonomous systems. How does this work, how common is it, and who does this? Jared Smith joins us on this episode of the Hedge to discuss the technique, and his research into how frequently it is used.

download

It Should Be Easy to Upgrade Your Load Balancer —And it Can Be

Applications have never been more important in business than they are today. And where there are applications, there’s a load balancer, working behind the scenes to ensure your applications can be used comfortably and safely at all times. When operating a load balancer, the most troublesome issue is upgrade work. Let’s examine the problems of traditional load balancer upgrades and take a look at VMware’s automated, streamlined solution: NSX Advanced Load Balancer.

Why do you need to upgrade your load balancer in the first place?

The main reasons to upgrade a load balancer are to patch vulnerabilities and bugs, to enable new features, and for EoSL support. A load balancer, located between users and applications, must above all be stable; we particularly want to avoid service disruptions due to defects in the load balancer software. For this reason, load balancer upgrades are inevitable. And for IT, the trick is to make them, as transparent and painless as possible..

For illustration, let’s take a look at a recent international case involving load balancers. To address software glitches, a project was running to upgrade hundreds of load balancers. The upgrade work was carried out little by little over several months, with operations personnel setting Continue reading

IBM moves toward consumption-based mainframe pricing

IBM continues to tweak its venerable mainframe to keep the Big Iron among the talking points in hybrid cloud.About a year ago the company changed its 20-year mainframe software pricing scheme to make it more palatable to hybrid cloud and multicloud users who might be thinking of moving workloads off the mainframe and into the cloud.To read this article in full, please click here

IBM moves toward consumption-based mainframe pricing

IBM continues to tweak its venerable mainframe to keep the Big Iron among the talking points in hybrid cloud.About a year ago the company changed its 20-year mainframe software pricing scheme to make it more palatable to hybrid cloud and multicloud users who might be thinking of moving workloads off the mainframe and into the cloud.To read this article in full, please click here

Prisma Access 2.0 Enables Painless Migration From Hardware Web Proxies

This guest post is by Drew Conry-Murray on behalf of Palo Alto Networks. We thank Palo Alto Networks for being a sponsor. Prisma Access, which delivers security services via the cloud, has added an explicit proxy feature in the 2.0 version. This feature can help companies migrate off of hardware-based proxies while still protecting Web […]

The post Prisma Access 2.0 Enables Painless Migration From Hardware Web Proxies appeared first on Packet Pushers.

IPv6 Buzz 075: Why Wells Fargo Bought Into IPv6

In this week's IPv6 Buzz episode, Ed, Scott, and Tom chat with John Burns, a lead architect at Wells Fargo, about the relatively early adoption of IPv6 at the company. The discussion also covers adoption trends in the financial sector as a whole, along with the key challenges and opportunities of the protocol.

The post IPv6 Buzz 075: Why Wells Fargo Bought Into IPv6 appeared first on Packet Pushers.

Risks and Rewards of the U.S. Broadband Funding Boom

The COVID-19 pandemic has taught us once and for all that broadband access is critical infrastructure. Without it, communities cannot work, learn, or earn online – a necessity during stay-at-home orders. And policymakers are taking notice. In the past few months, trillions of dollars have been proposed by the House, Senate, and White House for […]

The post Risks and Rewards of the U.S. Broadband Funding Boom appeared first on Internet Society.

Getting The Cloud But Keeping Control

The public cloud for years has tempted enterprises with the promise of much-needed agility and scalability that come with an elastic IT environment and of cost savings from not having to invest a lot of money upfront to buy a lot of infrastructure, adopting instead more flexible consumption models that allow organizations to pay only for what they use. …

Getting The Cloud But Keeping Control was written by Jeffrey Burt at The Next Platform.

Branch predictor: How many “if”s are too many? Including x86 and M1 benchmarks!

Branch predictor: How many
Branch predictor: How many

Some time ago I was looking at a hot section in our code and I saw this:


	if (debug) {
    	  log("...");
    }

This got me thinking. This code is in a performance critical loop and it looks like a waste - we never run with the "debug" flag enabled[1]. Is it ok to have if clauses that will basically never be run? Surely, there must be some performance cost to that...

Just how bad is peppering the code with avoidable if statements?

Back in the days the general rule was: a fully predictable branch has close to zero CPU cost.

To what extent is this true? If one branch is fine, then how about ten? A hundred? A thousand? When does adding one more if statement become a bad idea?

At some point the negligible cost of simple branch instructions surely adds up to a significant amount. As another example, a colleague of mine found this snippet in our production code:


const char *getCountry(int cc) {
		if(cc == 1) return "A1";
        if(cc == 2) return "A2";
        if(cc == 3) return "O1";
        if(cc == 4) return "AD";
        if(cc == 5) return "AE";
        if(cc == 6) return "AF";
         Continue reading

Nokia Lab | LAB 7 RSVP-TE Resource reservation |


Hello!

We're going ahead with constraint-based routing and today let's take a look into one of them in more detail - reservation of bandwidth resources.
I'm using topology and configuration from the previous lab.
Please check my first lab for input information.

Topology example

Lab tasks and questions:
  • Signaling and Reserving Bandwidth Requirements
  • create LSP from R1 to R6. The primary path should have bandwidth constraint (e.g. 500Mbit/s)
  • describe reserving bandwidth process
  • examine signaling with cspf and no cspf option
  • examine opaque LSA
    • check maximum bandwidth, reservable bandwidth, and unreserved bandwidth fields
    • Any changes after LSP signaling?
    • change path bandwidth and check opaque LSA again. Pay attention to Age and Sequence especially. What is a problem that can occur if we have an unstable network and a lot of LSP with bandwidth constraints?
    • How can we decrease the amount of LSA flood?
    • configure Threshold-Triggered IGP TE Updates and examine how it works
  • Bandwidth Reservation Styles
    • configure LSP to_R6 with primary "totally loose" path (bandwidth 200Mbit/s) and standby secondary "totally loose" path (bandwidth 300Mbit/s)
    • find a shared link
    • examine TED
    • What is unreserved bandwidth?
    • What is the default Bandwidth Reservation Style?
    • change Bandwidth Reservation Style and examine TED again
  • Least-Fill Bandwidth Reservation
    • 1 … 702 703 704 705 706 … 3,808