The Fat Pipe - All Packet Pushers Pods

The following text is a transcript of the audio 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 28, 2018, and here’s what’s happening. I had a briefing with DriveScale last month.

DriveScale is a composable infrastructure company, with a primary use case of distributed storage. That is, DriveScale software de-couples servers and storage, making it possible to compose infrastructure flexibly from pools of available resources. If that sounds more or less like what you get with public cloud, DriveScale claims to provide “the agility of the public cloud at a fraction of the cost.” That’s a plausible claim, as we’ve all learned that a company doesn’t move to the public cloud to save money.

I’ve written and spoken about DriveScale before, so search a bit if you’d like more detailed background on the company or check out the links in the episode notes. Now let’s jump into the discussion.

* An Introduction To DriveScale

* DriveScale’s Software Composable Infrastructure For Flash

DriveScale–Not A Hardware Company

In this briefing, DriveScale explained the current state of their evolution. One emphasis was that DriveScale wants to be out of the hardware business entirely. They are a software company that will work with a variety of different hardware. DriveScale will still supply some hardware if really needed to fill in gaps their customers might have, but that’s not what they want to be doing long-term.

Demonstrating how keen they are to be out of the hardware game, DriveScale highlighted some of their new industry alliances. For example, DriveScale is now a Dell EMC Tier 1 Enterprise Infrastructure Global Partner. That means Dell sales reps can sell DriveScale directly and get comped for it. That also means that if Dell EMC is an approved vendor in your shop, purchasing DriveScale is relatively easy for you.

DriveScale also announced support for the Cisco UCS S3260 Disk System, a monstrous series of disk chassis you attach to the network at high speed via dual QSFPs.

You get the idea. The more mainstream hardware DriveScale works with, the less DriveScale needs to supply their own hardware to make their solution saddle up a unicorn and ride around.

DriveScale Composer Architecture

Very well. DriveScale is all about the software. And what is the software? Much of the magic happens in DriveScale Composer. The chief value of DriveScale Composer is to compose any compute to any disk or flash, in a scalable way. DriveScale has successfully tested 3,000 nodes and 100,000 disks. 10,000 nodes and 100,000 disks are expected over time.

The system architecture is, of course, secure multi-tenant across the shared infrastructure, as you’d expect for something that scales so large. If you have a vast pool of resources, it’s safe to assume that a variety of discrete user groups would be consuming them. So, multi-tenancy is a necessary box DriveScale is checking here.

If you put multi-tenancy, compute, storage, and composability together, you might be wondering if DriveScale is a full-blown orchestrator, like Kubernetes. They are not. DriveScale Composer is RESTful API driven, and can be integrated with orchestrators. But DriveScale made it clear that they are not trying to orchestrate anyone’s applications themselves. Compose the infrastructure an application will use? Yes. Integrate that composable infrastructure with an orchestrator? Yes. Take on full orchestration duties themselves? No.

In addition to APIs, the DriveScale system leverages agents. These are user level agents that manage the resource...

The following text is a transcript of the audio 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 28, 2018, and here’s what’s happening. I had a briefing with DriveScale last month.

DriveScale is a composable infrastructure company, with a primary use case of distributed storage. That is, DriveScale software de-couples servers and storage, making it possible to compose infrastructure flexibly from pools of available resources. If that sounds more or less like what you get with public cloud, DriveScale claims to provide “the agility of the public cloud at a fraction of the cost.” That’s a plausible claim, as we’ve all learned that a company doesn’t move to the public cloud to save money.

I’ve written and spoken about DriveScale before, so search a bit if you’d like more detailed background on the company or check out the links in the episode notes. Now let’s jump into the discussion.

* An Introduction To DriveScale

* DriveScale’s Software Composable Infrastructure For Flash

DriveScale–Not A Hardware Company

In this briefing, DriveScale explained the current state of their evolution. One emphasis was that DriveScale wants to be out of the hardware business entirely. They are a software company that will work with a variety of different hardware. DriveScale will still supply some hardware if really needed to fill in gaps their customers might have, but that’s not what they want to be doing long-term.

Demonstrating how keen they are to be out of the hardware game, DriveScale highlighted some of their new industry alliances. For example, DriveScale is now a Dell EMC Tier 1 Enterprise Infrastructure Global Partner. That means Dell sales reps can sell DriveScale directly and get comped for it. That also means that if Dell EMC is an approved vendor in your shop, purchasing DriveScale is relatively easy for you.

DriveScale also announced support for the Cisco UCS S3260 Disk System, a monstrous series of disk chassis you attach to the network at high speed via dual QSFPs.

You get the idea. The more mainstream hardware DriveScale works with, the less DriveScale needs to supply their own hardware to make their solution saddle up a unicorn and ride around.

DriveScale Composer Architecture

Very well. DriveScale is all about the software. And what is the software? Much of the magic happens in DriveScale Composer. The chief value of DriveScale Composer is to compose any compute to any disk or flash, in a scalable way. DriveScale has successfully tested 3,000 nodes and 100,000 disks. 10,000 nodes and 100,000 disks are expected over time.

The system architecture is, of course, secure multi-tenant across the shared infrastructure, as you’d expect for something that scales so large. If you have a vast pool of resources, it’s safe to assume that a variety of discrete user groups would be consuming them. So, multi-tenancy is a necessary box DriveScale is checking here.

If you put multi-tenancy, compute, storage, and composability together, you might be wondering if DriveScale is a full-blown orchestrator, like Kubernetes. They are not. DriveScale Composer is RESTful API driven, and can be integrated with orchestrators. But DriveScale made it clear that they are not trying to orchestrate anyone’s applications themselves. Compose the infrastructure an application will use? Yes. Integrate that composable infrastructure with an orchestrator? Yes. Take on full orchestration duties themselves? No.

In addition to APIs, the DriveScale system leverages agents. These are user level agents that manage the resource...

The following text is a transcript of the audio 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 27, 2018, and here’s what’s happening. I had a briefing with HammerSpace last month. HammerSpace is a newly launched company to help you manage your data in shops where you are functioning both in the public cloud and on-premises.

If you said, “Oh, so HammerSpace is a storage company,” you’d be wrong about that. HammerSpace is about managing your data, no matter where it’s stored. In fact, a big part of the HammerSpace value proposition is that you don’t have to care about where the data is stored anymore. You can just get on with using it.

In this briefing, HammerSpace introduced their company, discussed some use cases, and dove deeply into their technology.

Understanding HammerSpace

For storage admins, the simplest way to describe HammerSpace is that they are an abstraction layer between data and metadata. David Flynn, CEO described it like this. “We have introduced an abstraction which separates the data consumer side and the storage infrastructure side. And that’s really what’s necessary before we can manage data in the cloud.”

HammerSpace becomes your point of entry to the data in your organization. HammerSpace indexes all the metadata, and presents a file system view via NFS or SMB. HammerSpace also manages access to the data, based on the policy that’s been defined.

They described this approach as sort of like DNS. HammerSpace clients talk to the metadata service, obtaining routing information for the file needed. The client then accesses the file directly. This means that HammerSpace’s metadata control plane is not in the data path. Storage platforms and data consumers no longer have to be tightly coupled in this model, as HammerSpace is the abstraction layer in the middle.

The Parallel NFS Link

If you’re wondering how HammerSpace gets out of the data path, the answer is parallel NFS. Quoting from pnfs.com, “Parallel NFS removes the performance bottleneck in traditional NAS systems by allowing the compute clients to read and write data directly and in parallel, to and from the physical storage devices. The NFS server is used only to control metadata and coordinate access, allowing incredibly fast access to very large data sets from many clients.”

HammerSpace is using this technology, and even has Trond Myklebust, Chief Linux Kernel Maintainer for NFS on the HammerSpace team as CTO.

For clients without parallel NFS support, HammerSpace offers a Data Service Node to act as a proxy.

The Automated, Policy-Governed Data Mover

All of this technology is lovely, but you might be wondering how this helps with hybrid cloud operations. The other major piece of the HammerSpace puzzle is that HammerSpace can move data around for you, wherever you need it, without operator intervention beyond policy definition.

David Flynn explained it this way. “Once you’re able to present data through the lens of the metadata and manage it through the lens of the metadata, then you can think of it as truly, ‘Data like air. Simply everywhere.’ As a matter of fact, the metadata can be aggressively replicated across the globe at different data centers so that you can have the view of it and when you go to access it, it can get the data to where you need it in a on-demand basis.”

HammerSpace Use Cases

HammerSpace cited “Cloud Analytics With On-Premises Data” as their main use case currently. The idea is that you have data in-house, and you need to mine that data. However, you want to use a cutting-edge cloud-based tool to do the analysis. Uh-oh. Normally,

The following text is a transcript of the audio 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 27, 2018, and here’s what’s happening. I had a briefing with HammerSpace last month. HammerSpace is a newly launched company to help you manage your data in shops where you are functioning both in the public cloud and on-premises.

If you said, “Oh, so HammerSpace is a storage company,” you’d be wrong about that. HammerSpace is about managing your data, no matter where it’s stored. In fact, a big part of the HammerSpace value proposition is that you don’t have to care about where the data is stored anymore. You can just get on with using it.

In this briefing, HammerSpace introduced their company, discussed some use cases, and dove deeply into their technology.

Understanding HammerSpace

For storage admins, the simplest way to describe HammerSpace is that they are an abstraction layer between data and metadata. David Flynn, CEO described it like this. “We have introduced an abstraction which separates the data consumer side and the storage infrastructure side. And that’s really what’s necessary before we can manage data in the cloud.”

HammerSpace becomes your point of entry to the data in your organization. HammerSpace indexes all the metadata, and presents a file system view via NFS or SMB. HammerSpace also manages access to the data, based on the policy that’s been defined.

They described this approach as sort of like DNS. HammerSpace clients talk to the metadata service, obtaining routing information for the file needed. The client then accesses the file directly. This means that HammerSpace’s metadata control plane is not in the data path. Storage platforms and data consumers no longer have to be tightly coupled in this model, as HammerSpace is the abstraction layer in the middle.

The Parallel NFS Link

If you’re wondering how HammerSpace gets out of the data path, the answer is parallel NFS. Quoting from pnfs.com, “Parallel NFS removes the performance bottleneck in traditional NAS systems by allowing the compute clients to read and write data directly and in parallel, to and from the physical storage devices. The NFS server is used only to control metadata and coordinate access, allowing incredibly fast access to very large data sets from many clients.”

HammerSpace is using this technology, and even has Trond Myklebust, Chief Linux Kernel Maintainer for NFS on the HammerSpace team as CTO.

For clients without parallel NFS support, HammerSpace offers a Data Service Node to act as a proxy.

The Automated, Policy-Governed Data Mover

All of this technology is lovely, but you might be wondering how this helps with hybrid cloud operations. The other major piece of the HammerSpace puzzle is that HammerSpace can move data around for you, wherever you need it, without operator intervention beyond policy definition.

David Flynn explained it this way. “Once you’re able to present data through the lens of the metadata and manage it through the lens of the metadata, then you can think of it as truly, ‘Data like air. Simply everywhere.’ As a matter of fact, the metadata can be aggressively replicated across the globe at different data centers so that you can have the view of it and when you go to access it, it can get the data to where you need it in a on-demand basis.”

HammerSpace Use Cases

HammerSpace cited “Cloud Analytics With On-Premises Data” as their main use case currently. The idea is that you have data in-house, and you need to mine that data. However, you want to use a cutting-edge cloud-based tool to do the analysis. Uh-oh. Normally,

On today's Network Break we get more insight into recent Cisco news on SD-WAN and WLAN controllers, discuss Google Cloud's curious choice for a new boss, examine the U.S. government's latest move against Huawei, and other tech news.

The following is a transcript of the audio recording 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 26, 2018, and here’s what’s happening. I had a briefing with Dell EMC last month. Dell EMC is one of the largest storage companies in the world. In this briefing, they focused on their Cyber Recovery 18.1 product.

You might be thinking, “Oh, another backup product. I already have one of those.” Sort of. Cyber Recovery is more than simply backup, and it’s more than what a decent disaster recovery plan gets you. Dell EMC points out that backup recovery is good, but can be quite slow if you need to recover a huge dataset, especially when dealing with tape. In addition, having a DR plan doesn’t mean you have a plan for recovering from a cyber attack. Disaster recovery and cyber attack recovery are different animals.

What do we mean by “cyber attack”? Dell EMC cited the infamous Petya and related attacks like NotPetya. These nasty bits of work are known as ransomware. Ransomware comes in various flavors, but in the Petya world encrypts files until you pay up while also attempting to harvest credentials and move through the network exploiting unpatched vulnerabilities.

Why is Dell EMC’s Cyber Recovery suited to address these attacks? The Cyber Recover Vault is an orchestrated Data Domain storage platform that provides an isolated copy of known good data that can be used to recover from a security breach. Let’s look more closely at each of the keywords in that definition, as that will help us understand exactly what Cyber Recovery is.

* First, orchestrated. The Cyber Recovery system is, in part, a bunch of software that does the logistical heavy lifting of deciding what data is known to be good and making sure there’s a copy of it in the vault.

* Second, Data Domain. Data Domain is an existing Dell EMC data protection product that’s got a pretty large install base. You need to be a Data Domain customer to use Dell EMC Cyber Recovery. Cyber Recovery leverages the ecosystem of apps and security features around Data Domain to help it do what it does and is replicating data from the Data Domain storage you’ve already got.

* Third, isolated. A big part of the security value of the Cyber Recovery Vault is that it is air-gapped. That is, the vault is only online when a data replication is in progress. To perform a replication, a network interface must be brought up. Then, a trusted connection between the two environments is created. Replication runs. Then the link connecting the vault to the rest of the network is shut down, creating a virtual air-gap between the vault and the rest of the network. In that way, if your network comes under an attack, the vault should be impervious, leaving you with a known good copy of your data.

* Fourth, known good. The data backed up into the vault is scanned to be sure it is good data before it is sent to the vault. Dell EMC claims that this takes the guesswork out of the restore process. The contents of the Cyber Recovery Vault are supposed to be completely trustworthy. That bit of magic is done by CyberSense, supplied to Dell EMC by a company called Index Engines. CyberSense works by scanning unstructured data and databases (Oracle, DB2, and SQL specifically) and scoring them based on an entropy engine measuring 40 different statistics. Machine learning is run to generate the score; there are no signatures involved. In effect, CyberSense is making a highly educated guess that something bad happened to a file based on the entropy score, and then replacing it with a known good copy. And that’s how you know you’ve got good files…

In today’s sponsored podcast we discuss the new demands that 5G will put on service providers, and how Cisco’s 5G xHaul Transport solution can meet those demands.

5G will create new demands on service provider networks. The existing MPLS network infrastructures of service providers won’t be adequate by themselves.

In today’s sponsored episode, we discuss exactly what 5G is asking of the network. We also examine how Cisco’s 5G xHaul Transport solution can meet those requirements.

Our guests are Simon Spraggs, Distinguished Consulting Engineer; and Dennis Hagarty, Technical Marketing Engineer with Cisco Systems.

We cover the technologies that 5G is bringing into service provider networks, including virtualization functions, network slicing, edge computing, and IoT.

Our guests explain why existing MPLS cores can’t handle 5G demands, and explain the main requirements for 5G networks.

We also look at how Cisco’s Converged xHaul Transport can help service providers address their 5G challenges and requirements.

Show Links:

Cisco Converged 5G xHaul Transport – Cisco Systems (Whitepaper)

Today's Priority Queue is a primer on satellite communications for military and commercial networks, including key basics on frequency bands and coverage areas, how different orbits affect network design, and more. Our guests are PC Drew and Evander Cook.

In Priority Queue 154, we chatted with PC Drew about the challenges of standing up a fully operational network in the middle of nowhere.

That’s standard operating procedure for the military as they fulfill their combat and humanitarian missions around the world.

In this episode, we’re going to go deeper on one facet of that conversation, satellite communications. PC Drew is back, and he brought along a colleague, Evander Cook, to educate us on SATCOM.

PC is Chief Technology Officer at SchoolBlocks, and Evander is Director of Network Operations at the United States Marine Corps.

We discuss basic concepts including frequency bands, TDMA vs. FDMA, satellite coverage areas, and the relationship between signal quality and signal power.

PC and Evander also explain how different satellite orbits affect network design, the different kinds of military and commercial services available, and how to account for moving vehicles in satellite communications.

Sponsor: Packet Pushers Virtual Design Clinic

Join the Packet Pushers for our next Virtual Design Clinic on December 19th, 2018. This live, free virtual event includes presentations on continuous integration and continuous testing for networking, and the impact of NVMe over Fabrics to network design. We’ll also have a panel of experts to partipate in“ask me anything” segments. Celebrate the holiday change freeze by registering at packetpushers.net/vdc.

Show Links:

PQ 154: Design Challenges In Military Networks – Packet Pushers

Orbital Altitudes of many significant satellites of earth – Wikipedia (jpg)