Enterprise-grade OpenStack

100% OpenStack, Enterprise-ready

Open Source platforms like OpenStack are proving to be the best solution for the demanding networking, data storage, compute horsepower and secure distributed infrastructure needs of enterprises and service providers. Open Cloud System (OCS) – with OpenStack at its core – delivers all of these essential capabilities in a completely open, integrated and massively scalable system that’s production-ready. Additionally, OCS interoperates natively with AWS and Google Compute Platform to deliver the fastest path to hybrid clouds.

With OCS 3.0, you can dramatically reduce OpenStack’s complexity and restore operational simplicity in your cloud-powered data center. We take core OpenStack, harden it and bake in key enterprise requirements that make it easier for IT and DevOPs to consume.

The 6 requirements of enterprise-grade OpenStack include:

1.99.99999% Uptime Control Plane 2.Robust Management
3.Open Architecture 4.Hybrid Cloud Interoperability
5.Scalability 6.Global Support and Services

By combining open source software, high-performance system design, comprehensive management tools and flexible hardware support, Cloudscaling delivers an unparalleled OpenStack experience. We ship only the most stable and hardened version of OpenStack with components that are tested, integrated and production-ready.

Cloudscaling is a Gold Member of the OpenStack Foundation and OCS 3.0 is licensed as an official distribution.

Cloud Operating System

Powering Web-scale IT

OCS 3.0 features a highly advanced Cloud Operating System that fully integrates an intelligent compute model with flexible storage options and hybrid networking.

OCS delivers a compute layer that intelligently places new VMs into your cloud infrastructure with the assurance they will perform with optimal efficiency and utilization, thereby lowering per-VM costs. This is made possible by OCS’s advanced scheduler that not only ensures performance guarantees, but has the ability to assign hardware designed for specific instance types.

Support for multiple block, object and fixed storage options gives you complete flexibility in choosing solutions that best fit your unique requirements. With BPS plug-ins for Cinder, you have the assurance of connecting to your existing enterprise storage vendors with the added benefit of AWS interoperability.

OCS ships with L3 networking to support a broad range of deployment options. Both leverage the same elastic networking architecture as modern public clouds such as AWS and GCE, employing high-speed L3 (layer-3) switch fabrics with 100% IP routing end-to-end to achieve maximum throughput at the lowest cost.

High Availability

99.999% Uptime Control Plane – there is no down time. Your cloud is always available.

Given the business criticality of your cloud computing infrastructure, uptime is among your top priorities. That’s why OCS features a powerful scale-up control plane that achieves “five nines” system availability.

This is really the place to distinguish between the data plane at 99.5% uptime and the control plane at 99.999% uptime, to bring up applications that “route around failure” and to tie into their requirements → cheap infrastructure (data plane) on demand coupled with an extreme uptime availability for the control plane; without these two one of these new apps can’t really succeed.

Elastic Cloud Reference Architecture

Your Private Cloud Blueprint

From our work in helping organizations across industries deploy OpenStack-powered private clouds, we have documented an elastic cloud reference architecture (ECRA) that can be used as the foundation for production deployments. ECRA combines a strategic organizational cloud model, QoS and SLA guarantees, a block-based data center approach and a failover design for so-called “cattle clouds”. ECRA provides an answer to the question of “What architecture do I go with” while offering a viable alternative to KVM-based architectures.

With ECRA providing an architectural “blueprint” approach, you’ll realize greater consistency across software releases, commonality among multi-cloud deployments and hybrid cloud interoperability – not just simple “compatibility”. In addition, you’ll realize faster time-to-market advantages and achieve much greater efficiency among your DevOps resources who no longer have to “reinvent the wheel” with each new cloud project.

Here’s what ECRA delivers in greater detail:

  • Cloud Organizational Model: Based on a “design-for-failure” approach, ECRA uses regions with a centralized control plane (APIs). This creates availability zones for fault isolation and nodes with elastic cloud resources. ECRA also provides multiple availability zones per region. A single set of services spans regions and availability zones for fault isolation.

    In addition, ECRA supports global services for users and billing; regional services for objects, load balancing and IP management; and availability zone services that include compute, block and networking.

  • Quality of Service & SLA Guarantees: ECRA is designed to support 99.5% data plane uptime (for elastic cloud resources) and 99.999% control plane uptime (for API endpoints). A design assumption is that elastic resources will fail at any time and that the app is responsible for managing these failures.

    To meet specific QoS and SLA guarantees, ECRA also provides sophisticated scheduling options. This includes bin-packing for optimal efficiency and utilization and dispersed risk mitigation of elastic cloud resource failure. Low latency resource clustering is also factored in for high performance.

  • Block-based Data Center Design: Using a “Lego-like” approach to building your cloud, ECRA is designed for scale and manageability. It advocates a standardized hardware design and bill of material model. In addition, ECRA recommends a separate and isolated control plane.

Hybrid Delivery

Full Public Cloud Interoperability

A true hybrid solution allows you to seamlessly move workloads across public and private environments to achieve greater scalability and flexibility for your applications. Cloudscaling delivers a private OpenStack-powered cloud that is fully interoperable with leading public cloud services such as Amazon Web Services (AWS) and Google Compute Platform. OCS is designed to fully support public cloud APIs, architectures, services and applications to harmonize clouds and provide a frictionless environment for app development and delivery.

Because OCS provides full API compatibility with both AWS and GCE, it can easily leverage common tools and libraries to shorten delivery cycles and accelerate time to market for cloud-native applications and services. API compatibility, however, is not enough. Behavioral fidelity is also required. OCS delivers a number of additional capabilities that provide a 1:1 mapping to key elastic cloud resources to recreate the architecture, cost model and scalability of supported public clouds, including:

  • AWS-style resource scheduling
  • Classic Networking in the EC2 model
  • Elastic Block Storage in the EC2 model
  • Block-based data center architecture

By minimizing differences in compute, storage, networking and APIs, OCS delivers the fastest path to hybrid clouds with OpenStack. IT benefits from increased cloud security and control while developers have the freedom to choose the cloud services that work best for a given workload.

Cloud Manager

Monitor and Control Your Infrastructure’s Performance

Quickly and effectively diagnosing performance issues in an OpenStack cloud can be challenging. OCS 3.0 now includes an intuitive interface that gives cloud operators a better way to see what’s going on in their cloud. Open Cloud Manager (OCM) sets the standard for managing enterprise-grade OpenStack by delivering granular real-time visibility into and control over every part of your private cloud. OCM empowers cloud operators to improve SLAs by identifying and resolving issues before they impact developers or the business.

OCM provides a cloud operations interface via graphical user interface (GUI), command line interface (CLI) and application programming interface (API) – to simplify systems management. OCM integrates with other logging tools for external syslog compatible events as well as monitoring and SIEM tools.

Key capabilities of OCM:

  • Monitor - maintain a central view of all activity in the control plane, including real-time data on storage capacity, network I/O, CPU usage and other key system metrics, as well as real time logs and events.
  • Diagnose - easily diagnose and resolve issues with operational intelligence, log viewing and search, system metrics graphs, audit trails and integration with Cloudscaling support.
  • Tune - adjust capacity in near real time based on usage, update OpenStack configurations and re-configure resource pools as required. You can also tune your cloud’s performance so that it meets required service level agreements by examining real time events via the log system, real time system metrics or using OpenStack resource pool management.
  • Manage – perform “cattle cloud management” to remove, reimage or install a server or proactively initiate repairs before they compromise business operations.

Hardware

Your Hardware, Your Data Center

We certify our solutions to run on hardware from Dell, H-P, Quanta and Super Micro. This allows you to choose hardware configurations from vetted reference architecture blueprints to meet your exacting workload and performance requirements. It also puts you in a position to evaluate more competitive pricing options.

Equally important, we deliver bare metal options as an elastic cloud service so your IT resources can manage equipment and processes they’re familiar with to deliver app performance that not only meets, but exceeds end user expectations.

In the Pets vs. Cattle debate, we think about hardware like cattle. If a cow gets ill, you can take it out back and lay it to rest. It doesn’t matter if one cow dies as long as the herd survives. For the modern data center that means commodity hardware takes center stage if you never want the infrastructure to slow down the pace at which your agile DevOps teams can operate. Cattle clouds really change the game in the data center – making software and the expertise that runs these machines more valuable.

OCS Runs Best on Cattle Clouds