Welcome!

Microservices Expo Authors: Stackify Blog, Aruna Ravichandran, Dalibor Siroky, Kevin Jackson, PagerDuty Blog

Related Topics: @CloudExpo, Java IoT, Microservices Expo, Cloud Security, @DXWorldExpo, SDN Journal

@CloudExpo: Blog Post

Traffic Advisory: Your Packets May Be Delayed

The past few years have seen a dramatic improvement in the latency in network switches

The past few years have seen a dramatic improvement in the latency in network switches. Single ASIC based switches can all pretty much switch packets in less than a microsecond. Current 10GE switching silicon provides anywhere from 300 to 800 nanoseconds, specialized silicon shaves that to less than 200 nanoseconds when limiting the amount of searching that needs to be done by reducing the size of lookup tables. Even other solutions play some smart tricks by providing forwarding hints for intermediate switches make those lookups take less than 50 nanoseconds.

Notice_Reduce_Speed_When_Entering_Parking_Lot_Signs_T41151-ba

Modular switches inherently have a higher latency. Line cards on modular switches typically have multiple ASICs, those ASICs are connected through a single or multi stage fabric. Each step takes time, resulting in latencies varying from around a microsecond when a packet stays on the same ASIC, to possibly 5-15 microseconds when a packet needs to travel through the fabric and back.

The speediest of ASICs achieve these low numbers by employing cut through switching. Cut through switching allows the ASIC to start transmitting a packet when enough of the header has been received to make a forwarding decision. The ASIC does not wait for the entire packet to be received (the more traditional store-and-forward mechanism), within the first few 100 bytes the forwarding decision has been made, and that same header (modified or not) is being transmitted out the destination port. It's somewhat odd to think that through, but the first bits of a packet may be received by the destination system before the last bits have left the first switch in the network.

Cut through switching comes with quite a few "buts". Most switches can only deploy cut through switching when the source and destination port are the same speed. 10GE in and 40GE out or vice versa is rarely supported and the ASIC will automatically switch to store-and-forward for those packets. For good reason. If a packet comes at you at 40GE rates, you simply cannot transmit it out a 10GE interface, that interface is not fast enough. In the reverse direction speed is not the issue, but if you were to employ cut through switching, for the duration of that packet your 40GE interface effectively runs at 10GE with lots of pauses in between pieces of a packet (figuratively speaking).

In addition, when the destination port has another packet being transmitted or in the queue, a new packet cannot be sent cut through. When another packet is ahead of you, you need to wait. And you may need to wait for quite a while. We often forget that it takes 1.2 microseconds to transmit a 1500 byte packet on a 10GE interface, more than 7 microseconds for a jumbo packet. When the destination port is being paused due to Data Center Bridging Priority Flow Control (PFC), the packet will be queued for store and forward. And make sure you add an extra 3 microseconds for 10GBASE-TX.

Datacenters are on a path to fewer layers of switching. Spine and leaf networks are being pitched as the best performing, low cost solution for dense networks. If you carefully examine the specs and pitches of some of the newer spine switches, you will notice that all of them make a case for deep buffers. Deep buffers assume that this switch needs to manage congestion by buffering packets, why else would you design expensive and power hungry buffer memory into those switches. Buffering and low latency don't go well together. If your spine and leaf network has nothing much to do, you may well see latency numbers of only a few microseconds or better. If the spine layer needs to buffer your packet, this number can jump up quickly to 10s of microseconds. And those large buffers seem to suggest it will.

There certainly are applications that are very sensitive to latency. Financial institution low latency trading networks are the example always used, and there are High Performance Computing environments with database, RDMI or similar applications that benefit from really low latency. Engineering the traffic in such a way that none of the low latency disruptive events described above happen is hard. Really hard. Extremely hard if there is a lot of traffic. Or a lot of endpoints. Networks that are specifically designed to aggregate and distribute (spine and leaf) will be more prone to these latency increasing scenarios. Creating a network with the ability to create isolated direct paths between switches that serve low latency applications is much more likely to avoid these. And even if the absolute latency is not the lowest, consistent latency with little jitter will certainly help the performance of adaptive mechanisms like TCP.

For the vast majority of applications in a typical enterprise datacenter, or a public cloud provider, the difference between 200 nanosecond switch latency and 1 microseconds is not measurably different in terms of user or application performance. The set of applications that behave noticably different at a few microseconds vs 10s of microseconds end to end latency is probably larger. I argue that you will get better results by carefully engineering the traffic for the applications that do care about low latency and low jitter. Make sure they get the bandwidth they need. Make sure they do not clash in the network with data hungry applications. Affinitize your network. It may just give you the latency and jitter performance you need.

The post Traffic advisory: your packets may be delayed appeared first on Plexxi.

Read the original blog entry...

More Stories By Marten Terpstra

Marten Terpstra is a Product Management Director at Plexxi Inc. Marten has extensive knowledge of the architecture, design, deployment and management of enterprise and carrier networks.

@MicroservicesExpo Stories
How is DevOps going within your organization? If you need some help measuring just how well it is going, we have prepared a list of some key DevOps metrics to track. These metrics can help you understand how your team is doing over time. The word DevOps means different things to different people. Some say it a culture and every vendor in the industry claims that their tools help with DevOps. Depending on how you define DevOps, some of these metrics may matter more or less to you and your team.
For many of us laboring in the fields of digital transformation, 2017 was a year of high-intensity work and high-reward achievement. So we’re looking forward to a little breather over the end-of-year holiday season. But we’re going to have to get right back on the Continuous Delivery bullet train in 2018. Markets move too fast and customer expectations elevate too precipitously for businesses to rest on their laurels. Here’s a DevOps “to-do list” for 2018 that should be priorities for anyone w...
If testing environments are constantly unavailable and affected by outages, release timelines will be affected. You can use three metrics to measure stability events for specific environments and plan around events that will affect your critical path to release.
In a recent post, titled “10 Surprising Facts About Cloud Computing and What It Really Is”, Zac Johnson highlighted some interesting facts about cloud computing in the SMB marketplace: Cloud Computing is up to 40 times more cost-effective for an SMB, compared to running its own IT system. 94% of SMBs have experienced security benefits in the cloud that they didn’t have with their on-premises service
DevOps failure is a touchy subject with some, because DevOps is typically perceived as a way to avoid failure. As a result, when you fail in a DevOps practice, the situation can seem almost hopeless. However, just as a fail-fast business approach, or the “fail and adjust sooner” methodology of Agile often proves, DevOps failures are actually a step in the right direction. They’re the first step toward learning from failures and turning your DevOps practice into one that will lead you toward even...
DevOps is under attack because developers don’t want to mess with infrastructure. They will happily own their code into production, but want to use platforms instead of raw automation. That’s changing the landscape that we understand as DevOps with both architecture concepts (CloudNative) and process redefinition (SRE). Rob Hirschfeld’s recent work in Kubernetes operations has led to the conclusion that containers and related platforms have changed the way we should be thinking about DevOps and...
While walking around the office I happened upon a relatively new employee dragging emails from his inbox into folders. I asked why and was told, “I’m just answering emails and getting stuff off my desk.” An empty inbox may be emotionally satisfying to look at, but in practice, you should never do it. Here’s why. I recently wrote a piece arguing that from a mathematical perspective, Messy Desks Are Perfectly Optimized. While it validated the genius of my friends with messy desks, it also gener...
The goal of Microservices is to improve software delivery speed and increase system safety as scale increases. Microservices being modular these are faster to change and enables an evolutionary architecture where systems can change, as the business needs change. Microservices can scale elastically and by being service oriented can enable APIs natively. Microservices also reduce implementation and release cycle time and enables continuous delivery. This paper provides a logical overview of the Mi...
The next XaaS is CICDaaS. Why? Because CICD saves developers a huge amount of time. CD is an especially great option for projects that require multiple and frequent contributions to be integrated. But… securing CICD best practices is an emerging, essential, yet little understood practice for DevOps teams and their Cloud Service Providers. The only way to get CICD to work in a highly secure environment takes collaboration, patience and persistence. Building CICD in the cloud requires rigorous ar...
The enterprise data storage marketplace is poised to become a battlefield. No longer the quiet backwater of cloud computing services, the focus of this global transition is now going from compute to storage. An overview of recent storage market history is needed to understand why this transition is important. Before 2007 and the birth of the cloud computing market we are witnessing today, the on-premise model hosted in large local data centers dominated enterprise storage. Key marketplace play...
The cloud revolution in enterprises has very clearly crossed the phase of proof-of-concepts into a truly mainstream adoption. One of most popular enterprise-wide initiatives currently going on are “cloud migration” programs of some kind or another. Finding business value for these programs is not hard to fathom – they include hyperelasticity in infrastructure consumption, subscription based models, and agility derived from rapid speed of deployment of applications. These factors will continue to...
Some people are directors, managers, and administrators. Others are disrupters. Eddie Webb (@edwardawebb) is an IT Disrupter for Software Development Platforms at Liberty Mutual and was a presenter at the 2016 All Day DevOps conference. His talk, Organically DevOps: Building Quality and Security into the Software Supply Chain at Liberty Mutual, looked at Liberty Mutual's transformation to Continuous Integration, Continuous Delivery, and DevOps. For a large, heavily regulated industry, this task ...
Following a tradition dating back to 2002 at ZapThink and continuing at Intellyx since 2014, it’s time for Intellyx’s annual predictions for the coming year. If you’re a long-time fan, you know we have a twist to the typical annual prediction post: we actually critique our predictions from the previous year. To make things even more interesting, Charlie and I switch off, judging the other’s predictions. And now that he’s been with Intellyx for more than a year, this Cortex represents my first ...
"Grape Up leverages Cloud Native technologies and helps companies build software using microservices, and work the DevOps agile way. We've been doing digital innovation for the last 12 years," explained Daniel Heckman, of Grape Up in this SYS-CON.tv interview at 21st Cloud Expo, held Oct 31 – Nov 2, 2017, at the Santa Clara Convention Center in Santa Clara, CA.
The Toyota Production System, a world-renowned production system is based on the "complete elimination of all waste". The "Toyota Way", grounded on continuous improvement dates to the 1860s. The methodology is widely proven to be successful yet there are still industries within and tangential to manufacturing struggling to adopt its core principles: Jidoka: a process should stop when an issue is identified prevents releasing defective products
We seem to run this cycle with every new technology that comes along. A good idea with practical applications is born, then both marketers and over-excited users start to declare it is the solution for all or our problems. Compliments of Gartner, we know it generally as “The Hype Cycle”, but each iteration is a little different. 2018’s flavor will be serverless computing, and by 2018, I mean starting now, but going most of next year, you’ll be sick of it. We are already seeing people write such...
Defining the term ‘monitoring’ is a difficult task considering the performance space has evolved significantly over the years. Lately, there has been a shift in the monitoring world, sparking a healthy debate regarding the definition and purpose of monitoring, through which a new term has emerged: observability. Some of that debate can be found in blogs by Charity Majors and Cindy Sridharan.
It’s “time to move on from DevOps and continuous delivery.” This was the provocative title of a recent article in ZDNet, in which Kelsey Hightower, staff developer advocate at Google Cloud Platform, suggested that “software shops should have put these concepts into action years ago.” Reading articles like this or listening to talks at most DevOps conferences might make you think that we’re entering a post-DevOps world. But vast numbers of organizations still struggle to start and drive transfo...
Let's do a visualization exercise. Imagine it's December 31, 2018, and you're ringing in the New Year with your friends and family. You think back on everything that you accomplished in the last year: your company's revenue is through the roof thanks to the success of your product, and you were promoted to Lead Developer. 2019 is poised to be an even bigger year for your company because you have the tools and insight to scale as quickly as demand requires. You're a happy human, and it's not just...
"Opsani helps the enterprise adopt containers, help them move their infrastructure into this modern world of DevOps, accelerate the delivery of new features into production, and really get them going on the container path," explained Ross Schibler, CEO of Opsani, and Peter Nickolov, CTO of Opsani, in this SYS-CON.tv interview at DevOps Summit at 21st Cloud Expo, held Oct 31 – Nov 2, 2017, at the Santa Clara Convention Center in Santa Clara, CA.