The Fat Pipe - All Packet Pushers Pods

The following is a transcript of the audio file you can listen to in the player above.

Welcome to Briefings In Brief, an audio digest of IT news and information from the Packet Pushers, including vendor briefings, industry research, and commentary.

I’m Ethan Banks, it’s November 19th, 2018, and here’s what’s happening:

* I had a briefing with Igneous earlier this month. Who’s Igneous? Igneous is focused on providing “as-a-service” solutions for unstructured data, including storage, backup & archival, global metadata indexing, and data workflow management. In this briefing, Igneous discussed recent announcements around their DataProtect, DataDiscover, and DataFlow services.

* Before we get to that, what is unstructured data? Unstructured data is, more or less, data that’s not in a highly structured form such what you’d find in a relational database. That doesn’t mean unstructured data doesn’t have any structure at all, but rather that the structure is not in a row/column format that lends itself well to carefully constructed queries. You know, like what SQL (structured query language) is all about.

* It’s the lack of nicely “queryable” structure that makes the unstructured data problem an interesting one to solve, as most data found at organizations is, in fact, unstructured. This is true even for big data shops, where high performance computing clusters run Hadoop and similar tools against massive unstructured datasets. In this unstructured data context, you find Igneous solving problems for companies. Let’s walk through the products they discussed with me in this briefing.

* First, DataProtect. DataProtect isn’t completely new, as it is the culmination of Igneous’ archive and backup offerings. The tool does what you’d expect. It’s about protecting your data in the form of backups, archival, and tiering of data to any public cloud, all delivered “as-a-service” meaning that Igneous is taking care of the DataProtect platform itself for you remotely.

* Second, DataDiscover. DataDiscover is about figuring out what’s interesting about your vast swaths of unstructured data by indexing all of the metadata that describes it. This service, by itself, is not about data movement or storage of the data. As the name suggests, DataDiscover wants to help you know what data you’ve got. Perhaps some of your data is old. Maybe there’s a huge volume of datasets, in which case DataDiscover helps you operate at scale. DataDiscover helps data managers understand what’s going in their data repositories and answer questions. Such as…is there data that should be archived? Where are the biggest files? Which files are the oldest? Etc.

* Third, DataFlow. DataFlow is about enabling end user workflows–making the lives of data consumers easier. In other words, this is less about IT folks who are storage professionals, and more about those who use the massive piles of data filling up all the storage volumes. The workflows around datasets tend to follow a lifecycle of capturing, processing, and analyzing data. Let’s give a simple high performance computing example. Imagine that, first, data is created by software. Then, the data is moved into an HPC cluster. The cluster runs software to process the data, generating more data in the form of results. That “results” dataset is copied to a directory for final analysis. Igneous DataFlow helps workflows around processes such as our little HPC example to be accomplished via APIs. The Igneous customer success team will work with the end user to make sure that metadata tagging, Python code, API calls, and overall workflow are functioning as well as possible. Again, remember that the DataFlow product is about the end user perspective. Igneous offered Paige.AI, a cancer research company, as a customer example for DataFlow…

The following is a transcript of the audio file you can listen to in the player above.

Welcome to Briefings In Brief, an audio digest of IT news and information from the Packet Pushers, including vendor briefings, industry research, and commentary.

I’m Ethan Banks, it’s November 19th, 2018, and here’s what’s happening:

* I had a briefing with Igneous earlier this month. Who’s Igneous? Igneous is focused on providing “as-a-service” solutions for unstructured data, including storage, backup & archival, global metadata indexing, and data workflow management. In this briefing, Igneous discussed recent announcements around their DataProtect, DataDiscover, and DataFlow services.

* Before we get to that, what is unstructured data? Unstructured data is, more or less, data that’s not in a highly structured form such what you’d find in a relational database. That doesn’t mean unstructured data doesn’t have any structure at all, but rather that the structure is not in a row/column format that lends itself well to carefully constructed queries. You know, like what SQL (structured query language) is all about.

* It’s the lack of nicely “queryable” structure that makes the unstructured data problem an interesting one to solve, as most data found at organizations is, in fact, unstructured. This is true even for big data shops, where high performance computing clusters run Hadoop and similar tools against massive unstructured datasets. In this unstructured data context, you find Igneous solving problems for companies. Let’s walk through the products they discussed with me in this briefing.

* First, DataProtect. DataProtect isn’t completely new, as it is the culmination of Igneous’ archive and backup offerings. The tool does what you’d expect. It’s about protecting your data in the form of backups, archival, and tiering of data to any public cloud, all delivered “as-a-service” meaning that Igneous is taking care of the DataProtect platform itself for you remotely.

* Second, DataDiscover. DataDiscover is about figuring out what’s interesting about your vast swaths of unstructured data by indexing all of the metadata that describes it. This service, by itself, is not about data movement or storage of the data. As the name suggests, DataDiscover wants to help you know what data you’ve got. Perhaps some of your data is old. Maybe there’s a huge volume of datasets, in which case DataDiscover helps you operate at scale. DataDiscover helps data managers understand what’s going in their data repositories and answer questions. Such as…is there data that should be archived? Where are the biggest files? Which files are the oldest? Etc.

* Third, DataFlow. DataFlow is about enabling end user workflows–making the lives of data consumers easier. In other words, this is less about IT folks who are storage professionals, and more about those who use the massive piles of data filling up all the storage volumes. The workflows around datasets tend to follow a lifecycle of capturing, processing, and analyzing data. Let’s give a simple high performance computing example. Imagine that, first, data is created by software. Then, the data is moved into an HPC cluster. The cluster runs software to process the data, generating more data in the form of results. That “results” dataset is copied to a directory for final analysis. Igneous DataFlow helps workflows around processes such as our little HPC example to be accomplished via APIs. The Igneous customer success team will work with the end user to make sure that metadata tagging, Python code, API calls, and overall workflow are functioning as well as possible. Again, remember that the DataFlow product is about the end user perspective. Igneous offered Paige.AI, a cancer research company, as a customer example for DataFlow…

Tech Bytes are short, sponsored conversations on the latest technologies and products. Today’s episode is sponsored by Silver Peak, and we talk to their customer First Bank.

Our guest is Marc Ashworth, VP – Information Security. We discuss how the bank was able to replace MPLS with lower-cost broadband connections, simplify its WAN edge by removing routers at each branch, and meet security requirements by moving to SD-WAN.

Ashworth shares his deployment experiences with swapping out MPLS with either broadband or cable modems at every branch.

We also discuss key features for First Bank, including WAN acceleration capabilities in Silver Peak, and encryption between branch devices and the bank’s headquarters.

Tech Bytes are short, sponsored conversations on the latest technologies and products. Today’s episode is sponsored by Silver Peak, and we talk to their customer First Bank.

Our guest is Marc Ashworth, VP – Information Security. We discuss how the bank was able to replace MPLS with lower-cost broadband connections, simplify its WAN edge by removing routers at each branch, and meet security requirements by moving to SD-WAN.

Ashworth shares his deployment experiences with swapping out MPLS with either broadband or cable modems at every branch.

We also discuss key features for First Bank, including WAN acceleration capabilities in Silver Peak, and encryption between branch devices and the bank’s headquarters.

On today's Network Break we examine why Oracle bought an SD-WAN vendor, discuss new Cisco switches and its new SD-WAN strategy, opine on BlackBerry's billion dollar security purchase, and more tech news.

The following is a transcription of the audio recording you can hear in the player above.

Welcome to Briefings In Brief, an audio digest of IT news and information from the Packet Pushers, including vendor briefings, industry research, and commentary.

I’m Ethan Banks, it’s November 15, 2018, and here’s what’s happening:

* I had a briefing with Oracle Cloud last month while attending a Tech Field Day event as a delegate.

* Oracle is, of course, one of the largest IT companies in the world. Their enterprise database, CRM, financial, and other applications have a massive install base that keep Oracle ticking along at $9 or $10 billion dollars in revenue or so per quarter.

* What you might not realize is that a significant percentage of that revenue comes from Oracle Cloud. That’s because what Oracle’s mostly done with their cloud is provide a home for enterprises to run Oracle applications. For shops who want to run their own workloads on Infrastructure as a Service, that might make Oracle Cloud seem uninteresting.

* However, there’s a lot more to the Oracle cloud story. Oracle Cloud Infrastructure is slowly working its way towards being feature competitive with the Big 3 – AWS, Azure, and GCP.

* How are they doing that? By building out robust cloud infrastructure aimed at the typical enterprise. Oracle’s take is that the Big 3 clouds aren’t meeting their customers’ needs. And again, when you hear “customers’ needs” in this context, you should think, “enterprise customers’ needs.”

* The claims here are that Big 3 cloud performance is inconsistent, and not well-suited for running Oracle apps. In addition, public cloud costs are high, sometimes with hidden or poorly understood charges. Add to that an inability to pick up and move when you want to if you’re leveraging PaaS, and maybe Oracle has a point.

* The focus for Oracle Cloud then is consistent, high performance, low, predictable pricing, and the ability to migrate enterprise workloads easily. In summary, making it easier for enterprises to operate in the cloud by giving them a cloud infrastructure that makes sense with the way typical enterprise applications are architected.

* This is an important point. Many shops that have attempted the “lift and shift” model of migrating to the cloud have been stunned by the resulting operational costs. That’s because the Big 3 aren’t setup to work as a mirror of your data center model. They are designed to be efficient when you use lots of tiny workloads spun up and shut down on demand as opposed to a few massive workloads that run 24×7. Think scale-out as opposed to scale-up, and maybe all the buzz about microservices starts to come into focus.

* So what do you do when re-architecting your applications isn’t really an option? Maybe, you talk to Oracle Cloud.

* Oracle Cloud Infrastructure offers, as you’d expect, compute, storage, and networking. There’s even a bare metal offering that’s a big as 52 cores, 8 GPUs, 768GB RAM, 51TB local SSD, and 5M IOPS according to my notes.

* Oracle cloud network struck me as a bit of unicorn magic. Rainbows all over the place. I didn’t get a deep-dive on how they are doing what they are doing, but Oracle has made some huge promises here. They describe the network as non-oversubscribed with a flat architecture, which I took to mean not leaf-spine, which is certainly possible. There are a lot of interesting network topologies out there aimed at specific traffic patterns. Whatever the topology is, Oracle will provide a cloud network performance SLA of less than < 100µs expected one way latency between hosts in an Availability Domain and < 500µs between Availability Domains. Oracle will provide credits if the performance falls below the SLA.

* Storage is NVMe SSDs everywhere, and importantly, Oracle does not …

The following is a transcription of the audio recording you can hear in the player above.

Welcome to Briefings In Brief, an audio digest of IT news and information from the Packet Pushers, including vendor briefings, industry research, and commentary.

I’m Ethan Banks, it’s November 15, 2018, and here’s what’s happening:

* I had a briefing with Oracle Cloud last month while attending a Tech Field Day event as a delegate.

* Oracle is, of course, one of the largest IT companies in the world. Their enterprise database, CRM, financial, and other applications have a massive install base that keep Oracle ticking along at $9 or $10 billion dollars in revenue or so per quarter.

* What you might not realize is that a significant percentage of that revenue comes from Oracle Cloud. That’s because what Oracle’s mostly done with their cloud is provide a home for enterprises to run Oracle applications. For shops who want to run their own workloads on Infrastructure as a Service, that might make Oracle Cloud seem uninteresting.

* However, there’s a lot more to the Oracle cloud story. Oracle Cloud Infrastructure is slowly working its way towards being feature competitive with the Big 3 – AWS, Azure, and GCP.

* How are they doing that? By building out robust cloud infrastructure aimed at the typical enterprise. Oracle’s take is that the Big 3 clouds aren’t meeting their customers’ needs. And again, when you hear “customers’ needs” in this context, you should think, “enterprise customers’ needs.”

* The claims here are that Big 3 cloud performance is inconsistent, and not well-suited for running Oracle apps. In addition, public cloud costs are high, sometimes with hidden or poorly understood charges. Add to that an inability to pick up and move when you want to if you’re leveraging PaaS, and maybe Oracle has a point.

* The focus for Oracle Cloud then is consistent, high performance, low, predictable pricing, and the ability to migrate enterprise workloads easily. In summary, making it easier for enterprises to operate in the cloud by giving them a cloud infrastructure that makes sense with the way typical enterprise applications are architected.

* This is an important point. Many shops that have attempted the “lift and shift” model of migrating to the cloud have been stunned by the resulting operational costs. That’s because the Big 3 aren’t setup to work as a mirror of your data center model. They are designed to be efficient when you use lots of tiny workloads spun up and shut down on demand as opposed to a few massive workloads that run 24×7. Think scale-out as opposed to scale-up, and maybe all the buzz about microservices starts to come into focus.

* So what do you do when re-architecting your applications isn’t really an option? Maybe, you talk to Oracle Cloud.

* Oracle Cloud Infrastructure offers, as you’d expect, compute, storage, and networking. There’s even a bare metal offering that’s a big as 52 cores, 8 GPUs, 768GB RAM, 51TB local SSD, and 5M IOPS according to my notes.

* Oracle cloud network struck me as a bit of unicorn magic. Rainbows all over the place. I didn’t get a deep-dive on how they are doing what they are doing, but Oracle has made some huge promises here. They describe the network as non-oversubscribed with a flat architecture, which I took to mean not leaf-spine, which is certainly possible. There are a lot of interesting network topologies out there aimed at specific traffic patterns. Whatever the topology is, Oracle will provide a cloud network performance SLA of less than < 100µs expected one way latency between hosts in an Availability Domain and < 500µs between Availability Domains. Oracle will provide credits if the performance falls below the SLA.

* Storage is NVMe SSDs everywhere, and importantly, Oracle does not …

Today's sponsored Weekly Show dives into Cisco's Network Assurance Engine. This software creates a real-time and continuously updated model of the network that IT can use to assess the impact of changes, get deep visibility into network state, and move network ops from reactive to proactive. We hear from an IT engineer inside Cisco who uses NAE in production.

Network engineers have lots of ways to officially measure how quickly they respond to changes or fix problems.

There’s also lots of informal ones: How soon can I go back to bed? Can I leave the office in time for dinner with the family? How long will it take to prove to the Dev team that no, it’s not the network?

On today’s Weekly Show we talk with our sponsor Cisco about its Network Assurance Engine (NAE), a software product designed to answer those important questions.

NAE uses a mathematical concept called formal verification to build a real-time model of the network. The model is derived from the continuous collection of each device’s state and configuration. This provides an accurate and dynamic representation of the network, as opposed to a static diagram or an engineer’s best recollection.

NAE uses this model to assess the impact of changes, prevent outages, get deep visibility into network state, and move network operations from a reactive to a proactive stance.

Our guests are Navneet Yadav, Sr. Director Product Management, Head of Product – Cisco Network Assurance Engine; and Vishal Soni, Sr. Engineer, Cisco IT DC Network.

We discuss how NAE works, the concepts of formal verification and how they’re applied to networking, how NAE improves Day 2 (and beyond) operations, and customer use cases.

We also get a real-world perspective from Vishal, who uses NAE in production in Cisco’s own internal IT organization. He’ll offer his perspective on the product, and share concrete details about NAE’s impact on operations including change management, troubleshooting, and incident handling.

Show Links:

Network Assurance – Cisco Systems

Cisco Network Assurance Engine (NAE): Customer testimonial by Cisco IT – Cisco Systems via YouTube

Verify, Or Die Trying: Observations on Change Management – Gestalt IT

Assure Network Security Policy and Compliance in the Data Center with Cisco Network Assurance Engine – Gestalt IT

Change Doesn’t Have To Be a Four Letter Word – Gestalt IT

Configuration and Hardware Assurance in the Datacenter with Cisco Network Assurance Engine – Gestalt IT

Hands On with Cisco Network Assurance Engine – Gestalt IT

Cisco Network Assurance Engine: From Download to Value in 60 Minutes (or less) – Gestalt IT

Networking Has Changed, Have You? – Gestalt IT

In today's IPv6 Buzz episode we dive into the challenges of IPv6 adoption in the enterprise. We discuss how and why to set up an internal IPv6 working group, why you need more than the networking team involved, and other key issues. Our guest is Silvia Hagen.