SimpliVity is pioneering the next generation of converged infrastructure, Convergence 3.0. This is the first in a series of posts in which we provide an overview of the convergence landscape and demonstrate how SimpliVity is able to provide benefits and customer value far beyond all other offerings in the market. Click here for my presentation at the Gartner Data Center Conference 2013 in which I introduced Convergence 3.0.
The IT infrastructure market is undergoing unprecedented transformation. The most significant transformation is reflected by two narratives: “Convergence” and “Software-Defined Data Center” (SDDC). Both are responses to today’s IT realities of infrastructure clutter, complexity, and high cost, and they represent attempts to simplify IT and to reduce overall infrastructure TCO. The figure below exemplifies the high cost and complexity of today’s infrastructure, which is typically comprised of at least 8-12 different hardware and software products from as many vendors, each with a different management interface, and each requiring different training.
Importantly, each of the separate products that comprise this legacy stack is grossly overprovisioned with its own resources (CPU, DRAM, Storage) in order to address intermittent peak workloads of the resident applications. The value of a single shared resource pool that was offered by server virtualization is still limited only to the server layer. All other products are islands of overprovisioned compute resources that are not shared, and therefore give rise to low utilization of the overall stack and therefore cause the ripple effects of high acquisition costs and high data center space and power. Simply put, too many compute resources are wasted in today’s legacy environments. The following diagram exemplifies the value of shared resource pools as applied in the server layer. Prior to Convergence 3.0, this concept had not been applied across all other products that make up the infrastructure stack.
Up until recently, both the Convergence and the SDDC narratives have mostly represented a promise of what’s to come, rather than a reality that customers can deploy in their data centers.
The first two generations of Convergence—Convergence 1.0 and Convergence 2.0—have delivered on a small subset of the comprehensive vision, but fall short of delivering the full value promised by the concept. While they have provided some value in terms of simplified management and reduced deployment time they’ve not comprehensively addressed the core underlying problems that drive today’s clutter and complexity.
For convergence to truly deliver on the promise, a new data architecture is required that includes and enables the following:
- Full convergence of all infrastructure functionality running in a single shared resource pool. This includes the specialty data deduplication and data protection appliances that have proliferated IT and are out of the scope of Convergence 1.0 and 2.0
- Novel data architecture that:
- Addresses IOPS problem: increases performance and reduces latency
- Addresses data mobility globally
- Reduces storage capacity and associated space and power
- Global federated management
SimpliVity is pioneering Convergence 3.0, the next generation of converged infrastructure which is fulfilling the originally sought benefits, powered by SimpliVity’s core innovative data architecture. Interestingly, complete realization of Convergence 3.0 is technologically the same as SDDC, though the packaging is different.
Convergence 1, 2, 3 – A Brief History
In 2009, in sensing the frustration of customers across the globe who faced the complications of implementing and managing bloated 20th century infrastructure, vendors began offering converged products that promised to deliver multiple layers of the IT stack in single solution. The sought benefits of convergence are clear: simplification of management, a well-organized and synthesized IT stack, reduction of vendor friction, and ultimately, cost savings.
Convergence 1.0 products addressed the outcry of end-users, who were desperately seeking core IT products in a single rack with common management. These products were stop-gap, engineered solutions delivered by incumbent vendors or a coalition of vendors that integrated existing products, including server, server virtualization, and storage and switch, under a newly developed common management framework. The figure below illustrates the Convergence 1.0 approach: simply bundling existing legacy products and delivering them as a single offering, sometimes with a new interface.
However, like the original legacy stack, these products leave a lot to be desired. They do not improve efficiency as there is no improvement to resource pooling. Further, they do not significantly improve on the data center complexity problem given that they do not include data protection, cloud, WAN optimization, or deduplication functionality, leaving the IT team to continue to source these solutions separately.
Convergence 2.0 products offer some IT infrastructure functionality on top of a single, shared x86 resource pool. These products bundle the primary stack—servers, virtualization, and storage—but do not address the broader scope of functionality of the data center and therefore do not go far enough in reducing complexity and cost.
While these products provide enhanced utilization and simplification of management, they don’t incorporate all the functionality needed to solve data bloat, and because of this, IOPS remain expensive and IOPS redundancy isn’t addressed. And without global data efficiency, there is no practical ability to manage geographically dispersed sites from a single management pane.
Convergence 3.0 represents the logical next step in the evolution of infrastructure convergence. Convergence 3.0 delivers radical simplification and TCO savings by consolidating all requisite infrastructure functionality into a single infrastructure stack, running on an efficient, elastic pool of x86 resources. In doing so, Convergence 3.0 delivers on the promise of SDDC at the technological level.
Both Convergence 3.0 and SDDC include the following:
- A single functional IT stack, that runs on commodity x86 resources and scales elastically, by virtue of adding resources (CPU, memory, flash, HDD, and network ports) at a ratio that the end-user desires
- All resources are virtualized, fungible, and available to the business or infrastructure applications, delivering the promise of a Single Shared Resource Pool.
- Full convergence of all IT infrastructure functions and avoidance of clutter, including:
- Core tenants of IT infrastructure: server, server virtualization, network, primary storage, backup storage, and archival storage
- All other requisite infrastructure functions for managing and protecting data:
- SSD-accelerated performance
- Storage caching
- VM centric backup and replication
- Deduplication-enabled disk-based backup
- WAN optimization
- Public cloud integration
- Global Federated Management of all IT sites and data centers from a single management point. This includes the ability to efficiently mobilize the data across sites.
SimpliVity is the first and only vendor delivering Convergence 3.0 today. Delivering Convergence 3.0 is not an easy task, so it should be no surprise that there is only one company delivering this today. It requires an entirely new and highly innovative data architecture that has been designed for high performance, scalability, maximum data efficiency, and global management. Highlights of SimpliVity’s data architecture include:
- All data is granularly deduped and compressed, inline, at the point of data creation, once and sustained forever, across all layers of storage media (including DRAM, SSD, and HDD)
- Improves performance, reduces latency, reducing IOPS (since redundant reads and writes are avoided!)
- Facilitates efficient data mobility globally, across geographically dispersed sites
- Reduces required storage capacity and associated data center space and power
- Globally addressable data that is aligned with VMs and accessible globally to any VM
- Global Federated Management of all nodes and all VMs across geographically distributed data centers and sites
Powered by this innovative data architecture, SimpliVity is able to deliver radical simplification by eliminating the need for all of the core infrastructure products and specialty appliances that today create infrastructure clutter and complexity, and force IT teams to spend most of their budget simply to maintain the status quo. In doing so SimpliVity is delivering on the originally promised benefits of Convergence and SDDC: IT simplification and TCO savings.
The table below summarizes the key differences between the three generations of converged infrastructure offerings, demonstrating that Convergence 3.0 is finally delivering on the vision of infrastructure convergence.
In subsequent posts we will provide further details on SimpliVity’s underlying technology and the impact it is making for customers across the globe. Until then, I encourage you to learn more: