Disaggregation and Composability vital for AI/DL models to scale

New generations of applications and workloads like AI/DL (Artificial Intelligence/Deep Learning), and HPC (High Performance Computing) are breaking the seams of entrenched storage infrastructure models and frameworks. We cannot continue to scale-up or scale-out the storage infrastructure to meet these inundating fluctuating I/O demands. It is time to look at another storage architecture type of infrastructure technology – Composable Infrastructure Architecture.

Infrastructure is changing. The previous staid infrastructure architecture parts of compute, network and storage have long been thrown of the window, precipitated by the rise of x86 server virtualization almost 20 years now. It triggered a tsunami of virtualizing everything, including storage virtualization, which eventually found a more current nomenclature – Software Defined Storage. Both storage virtualization and software defined storage (SDS) are similar and yet different and should be revered through different contexts and similar goals. This Tech Target article laid out both nicely.

As virtualization raged on, converged infrastructure (CI) which evolved into hyperconverged infrastructure (HCI) went fever pitch for a while. Companies like Maxta, Pivot3, Atlantis, are pretty much gone, with HPE® Simplivity and Cisco® Hyperflex occasionally blipped in my radar. In a market that matured very fast, HCI is now dominated by Nutanix™ and VMware®, with smaller Microsoft®, Dell EMC® following them.

From HCI, the attention of virtualization has shifted something more granular, more scalable in containerization. Despite a degree of complexity, containerization is taking agility and scalability to the next level. Kubernetes, Dockers are now mainstay nomenclature of infrastructure engineers and DevOps. So what is driving composable infrastructure? Have we reached the end of virtualization? Not really.

Evolution of infrastructure. Source: IDC

It is just that one part of the infrastructure landscape is changing. This new generation of AI/ML workloads are flipping the coin to the other side of virtualization. As we see the diagram above, IDC brought this mindset change to get us to Think Composability, the next phase of Infrastructure.

Continue reading

Project COSI

The S3 (Simple Storage Service) has become a de facto standard for accessing object storage. Many vendors claim 100% compatibility to S3, but from what I know, several file storage services integration and validation with the S3 have revealed otherwise. There are certain nuances that have derailed some of the more advanced integrations. I shall not reveal the ones that I know of, but let us use this thought as a basis of our discussion for Project COSI in this blog.

Project COSI high level architecture

What is Project COSI?

COSI stands for Container Object Storage Interface. It is still an alpha stage project in Kubernetes version 1.25 as of September 2022 whilst the latest version of Kubernetes today is version 1.26. To understand the objectives COSI, one must understand the journey and the challenges of persistent storage for containers and Kubernetes.

For me at least, there have been arduous arguments of provisioning a storage repository that keeps the data persistent (and permanent) after containers in a Kubernetes pod have stopped, or replicated to another cluster. And for now, many storage vendors in the industry have settled with the CSI (container storage interface) framework when it comes to data persistence using file-based and block-based storage. You can find a long list of CSI drivers here.

However, you would think that since object storage is the most native storage to containers and Kubernetes pods, there is already a consistent way to accessing object storage services. From the objectives set out by Project COSI, turns out that there isn’t a standard way to provision and accessing object storage as compared to the CSI framework for file-based and block-based storage. So the COSI objectives were set to:

  • Kubernetes Native – Use the Kubernetes API to provision, configure and manage buckets
  • Self Service – A clear delineation between administration and operations (DevOps) to enable self-service capability for DevOps personnel
  • Portability – Vendor neutrality enabled through portability across Kubernetes Clusters and across Object Storage vendors

Further details describing Project COSI can be found here at the Kubernetes site titled “Introducing COSI: Object Storage Management using Kubernetes API“.

Standardization equals technology adoption

Standardization means consistency, control, confidence. The higher the standardization across the storage and containerized apps industry, the higher the adoption of the technology. And given what I have heard from the industry over these few years, Kubernetes, to me, even till this day, is a platform and a framework that are filled and riddled with so many moving parts. Many of the components looks the same, feels the same, and sounds the same, but might not work out the same when deployed.

Therefore, the COSI standardization work is important and critical to grow this burgeoning segment, especially when we are rocketing towards disaggregation of computing service units, resources that be orchestrated to scale up or down at the execution of codes. Infrastructure-as-Code (IAC) is becoming a reality more and more with each passing day, and object storage is at the heart of this transformation for Kubernetes and containers.

Continue reading

Time to Conflate Storage with Data Services

Around the year 2016, I started to put together a better structure to explain storage infrastructure. I started using the word Data Services Platform before what it is today. And I formed a pictorial scaffold to depict what I wanted to share. This was what I made at that time.

Data Services Platform (circa 2016)- Copyright Heoh Chin Fah

One of the reasons I am bringing this up again is many of the end users and resellers still look at storage from the perspective of capacity, performance and price. And as if two plus two equals five, many storage pre-sales and architects reciprocate with the same type of responses that led to the deteriorated views of the storage technology infrastructure industry as a whole. This situation irks me. A lot.

Continue reading

Layers in Storage – For better or worse

Storage arrays and storage services are built upon by layers and layers beneath its architecture. The physical components of hard disk drives and solid states are abstracted into RAID volumes, virtualized into other storage constructs before they are exposed as shares/exports, LUNs or objects to the network.

Everyone in the storage networking industry, is cognizant of the layers and it is the foundation of knowledge and experience. The public cloud storage services side is the same, albeit more opaque. Nevertheless, both have layers.

In the early 2000s, SNIA® Technical Council outlined a blueprint of the SNIA® Shared Storage Model, a framework describing layers and properties of a storage system and its services. It was similar to the OSI 7-layer model for networking. The framework helped many industry professionals and practitioners shaped their understanding and the development of knowledge in their respective fields. The layering scheme of the SNIA® Shared Storage Model is shown below:

SNIA Shared Storage Model – The layering scheme

Storage vendors layering scheme

While SNIA® storage layers were generic and open, each storage vendor had their own proprietary implementation of storage layers. Some of these architectures are simple, but some, I find a bit too complex and convoluted.

Here is an example of the layers of the Automated Volume Management (AVM) architecture of the EMC® Celerra®.

EMC Celerra AVM Layering Scheme

I would often scratch my head about AVM. Disks were grouped into RAID groups, which are LUNs (Logical Unit Numbers). Then they were defined as Celerra® dvols (disk volumes), and stripes of the dvols were consolidated into a storage pool.

From the pool, a piece of a storage capacity construct, called a slice volume, were combined with other slice volumes into a metavolume which eventually was presented as a file system to the network and their respective NAS clients. Explaining this took an effort because I was the IP Storage product manager for EMC® between 2007 – 2009. It was a far cry from the simplicity of NetApp® ONTAP 7 architecture of RAID groups and volumes, and the WAFL® (Write Anywhere File Layout) filesystem.

Another complicated layered framework I often gripe about is Ceph. Here is a look of how the layers of CephFS is constructed.

Ceph Storage Layered Framework

I work with the OpenZFS filesystem a lot. It is something I am rather familiar with, and the layered structure of the ZFS filesystem is essentially simpler.

Storage architecture mixology

Engineers are bizarre when they get too creative. They have a can do attitude that transcends the boundaries of practicality sometimes, and boggles many minds. This is what happens when they have their own mixology ideas.

Recently I spoke to two magnanimous persons who had the idea of providing Ceph iSCSI LUNs to the ZFS filesystem in order to use the simplicity of NAS file sharing capabilities in TrueNAS® CORE. From their own words, Ceph NAS capabilities sucked. I had to draw their whole idea out in a Powerpoint and this is the architecture I got from the conversation.

There are 3 different storage subsystems here just to provide NAS. As if Ceph layers aren’t complicated enough, the iSCSI LUNs from Ceph are presented as Cinder volumes to the KVM hypervisor (or VMware® ESXi) through the Cinder driver. Cinder is the persistent storage volume subsystem of the Openstack® project. The Cinder volumes/hypervisor datastore are virtualized as vdisks to the respective VMs installed with TrueNAS® CORE and OpenZFS filesystem. From the TrueNAS® CORE, shares and exports are provisioned via the SMB and NFS protocols to Windows and Linux respectively.

It works! As I was told, it worked!

A.P.P.A.R.M.S.C. considerations

Continuing from the layered framework described above for NAS, other aspects beside the technical work have to be considered, even when it can work technically.

I often use a set of diligent data storage focal points when considering a good storage design and implementation. This is the A.P.P.A.R.M.S.C. Take for instance Protection as one of the points and snapshot is the technology to use.

Snapshots can be executed at the ZFS level on the TrueNAS® CORE subsystem. Snapshots can be trigged at the volume level in Openstack® subsystem and likewise, rbd snapshots at the Ceph subsystem. The question is, which snapshot at which storage subsystem is the most valuable to the operations and business? Do you run all 3 snapshots? How do you execute them in succession in a scheduled policy?

In terms of performance, can it truly maximize its potential? Can it churn out the best IOPS, and deliver at wire speed? What is the latency we can expect with so many layers from 3 different storage subsystems?

And supporting this said architecture would be a nightmare. Where do you even start the troubleshooting?

Those are just a few considerations and questions to think about when such a layered storage architecture along. IMHO, such a design was over-engineered. I was tempted to say “Just because you can, doesn’t mean you should

Elegance in Simplicity

Einstein (I think) quoted:

Einstein’s quote on simplicity and complexity

I am not saying that having too many layers is wrong. Having a heavily layered architecture works for many storage solutions out there, where they are often masked with a simple and intuitive UI. But in yours truly point of view, as a storage architecture enthusiast and connoisseur, there is beauty and elegance in simple designs.

The purpose here is to promote better understanding of the storage layers, and how they integrate and interact with each other to deliver the data services to the network. In the end, that is how most storage architectures are built.

 

Storage in a shiny multi-cloud space

The multi-cloud for infrastructure-as-a-service (IaaS) era is not here (yet). That is what the technology marketers want you to think. The hype, the vapourware, the frenzy. It is what they do. The same goes to technology analysts where they describe vision and futures, and the high level constructs and strategies to get there. The hype of multi-cloud is often thought of running applications and infrastructure services seamlessly in several public clouds such as Amazon AWS, Microsoft® Azure and Google Cloud Platform, and linking it to on-premises data centers and private clouds. Hybrid is the new black.

Multicloud connectivity to public cloud providers and on-premises private cloud

Multi-Cloud, on-premises, public and hybrid clouds

And the aspiration of multi-cloud is the right one, when it is truly ready. Gartner® wrote a high level article titled “Why Organizations Choose a Multicloud Strategy“. To take advantage of each individual cloud’s strengths and resiliency in respective geographies make good business sense, but there are many other considerations that cannot be an afterthought. In this blog, we look at a few of them from a data storage perspective.

In the beginning there was … 

For this storage dinosaur, data storage and compute have always coupled as one. In the mainframe DASD days. these 2 were together. Even with the rise of networking architectures and protocols, from IBM SNA, DECnet, Ethernet & TCP/IP, and Token Ring FC-SAN (sorry, this is just a joke), the SANs, the filers to the servers were close together, albeit with a network buffered layer.

A decade ago, when the public clouds started appearing, data storage and compute were mostly inseparable. There was demarcation of public clouds and private clouds. The notion of hybrid clouds meant public clouds and private clouds can intermix with on-premise computing and data storage but in almost all cases, this was confined to a single public cloud provider. Until these public cloud providers realized they were not able to entice the larger enterprises to move their IT out of their on-premises data centers to the cloud convincingly. So, these public cloud providers decided to reverse their strategy and peddled their cloud services back to on-prem. Today, Amazon AWS has Outposts; Microsoft® Azure has Arc; and Google Cloud Platform launched Anthos.

Continue reading

Persistent Storage could stifle Google Anthos multi-cloud ambitions

To win in the multi-cloud game, you have to be in your competitors’ cloud. Google Cloud has been doing that since they announced Google Anthos just over a year ago. They have been crafting their “assault”, starting with on-premises, and Anthos on AWS. Anthos on Microsoft® Azure is coming, currently in preview mode.

Google CEO Sundar Pichai announcing Google Anthos at Next ’19

BigQuery Omni conversation starter

2 weeks ago, whilst the Google Cloud BigQuery Omni announcement was still under wraps, local Malaysian IT portal Enterprise IT News sent me the embargoed article to seek my views and opinions. I have to admit that I was ignorant about the deeper workings of BigQuery, and haven’t fully gone through the works of Google Anthos as well. So I researched them.

Having done some small works on Qubida (defunct) and Talend several years ago, I have grasped useful data analytics and data enablement concepts, and so BigQuery fitted into my understanding of BigQuery Omni quite well. That triggered my interests to write this blog and meshing the persistent storage conundrum (at least for me it is something to be untangled) to Kubernetes, to GKE (Google Kubernetes Engine), and thus Anthos as well.

For discussion sake, here is an overview of BigQuery Omni.

An overview of Google Cloud BigQuery Omni on multiple cloud providers

My comments and views are in this EITN article “Google Cloud’s BigQuery Omni for Multi-cloud Analytics”.

Continue reading

Paradigm shift of Dev to Storage Ops

[ Disclosure: I was invited by GestaltIT as a delegate to their Storage Field Day 19 event from Jan 22-24, 2020 in the Silicon Valley USA. My expenses, travel, accommodation and conference fees were covered by GestaltIT, the organizer and I was not obligated to blog or promote the vendors’ technologies presented at the event. The content of this blog is of my own opinions and views ]

A funny photo (below) came up on my Facebook feed a couple of weeks back. In an honest way, it depicted how a developer would think (or the lack of thinking) about the storage infrastructure designs and models for the applications and workloads. This also reminded me of how DBAs used to diss storage engineers. “I don’t care about storage, as long as it is RAID 10“. That was aeons ago 😉

The world of developers and the world of infrastructure people are vastly different. Since cloud computing birthed, both worlds have collided and programmable infrastructure-as-code (IAC) have become part and parcel of cloud native applications. Of course, there is no denying that there is friction.

Welcome to DevOps!

The Kubernetes factor

Containerized applications are quickly defining the cloud native applications landscape. The container orchestration machinery has one dominant engine – Kubernetes.

In the world of software development and delivery, DevOps has taken a liking to containers. Containers make it easier to host and manage life-cycle of web applications inside the portable environment. It packages up application code other dependencies into building blocks to deliver consistency, efficiency, and productivity. To scale to a multi-applications, multi-cloud with th0usands and even tens of thousands of microservices in containers, the Kubernetes factor comes into play. Kubernetes handles tasks like auto-scaling, rolling deployment, computer resource, volume storage and much, much more, and it is designed to run on bare metal, in the data center, public cloud or even a hybrid cloud.

Continue reading

Is General Purpose Object Storage disenfranchised?

[Disclosure: I am invited by GestaltIT as a delegate to their Storage Field Day 19 event from Jan 22-24, 2020 in the Silicon Valley USA. My expenses, travel, accommodation and conference fees will be covered by GestaltIT, the organizer and I am not obligated to blog or promote the vendors’ technologies to be presented at this event. The content of this blog is of my own opinions and views]

This is NOT an advertisement for coloured balls.

This is the license to brag for the vendors in the next 2 weeks or so, as we approach the 2020 new year. This, of course, is the latest 2019 IDC Marketscape for Object-based Storage, released last week.

My object storage mentions

I have written extensively about Object Storage since 2011. With different angles and perspectives, here are some of them:

Continue reading

Figuring out storage for Kubernetes and containers

Oops! I forgot about you!

To me, containers and container orchestration (CO) engines such as Kubernetes, Mesos, Docker Swarm are fantastic. They scale effortlessly and are truly designed for cloud native applications (CNA).

But one thing irks me. Storage management for containers and COs. It was as if when they designed and constructed containers and the containers orchestration (CO) engines, they forgot about the considerations of storage and storage management. At least the persistent part of storage.

Over a year ago, I was in two minds about persistent storage, especially when it comes to the transient nature of microservices which was so prevalent and were inundating the cloud native applications landscape. I was searching for answers in my blog. The decentralization of microservices in containers means mass deployment at the edge, but to have the pre-processed and post-processed data stick to the persistent storage at the edge device is a challenge. The operative word here is “STICK”.

Two different worlds

Containers were initially designed and built for lightweight applications such as microservices. The runtime, libraries, configuration files and dependencies are all in one package. They were meant to do simple tasks quickly and scales to thousands easily. They could be brought up and brought down in little time and did not have to bother about the persistent data stored by the host. The state of the containers were also not important to the application tasks at hand.

Today containers like Docker have matured to run enterprise applications and the state of the container is important. The applications must know the state and the health of the container. The container could be in online mode, online but not accepting data mode, suspended mode, paused mode, interrupted mode, quiesced mode or halted mode. Each mode or state of the container is important to the running applications and the container can easily brought up or down in an instance of a command. The stateful nature of the containers and applications is critical for the business. The same situation applies to container orchestration engines such as Kubernetes.

Container and Kubernetes Storage

Docker provides 3 methods to local storage. In the diagram below, it describes:

Continue reading