As an international media and entertainment company, Scanline VFX provides full-service visual effects for feature films and commercials. The company began in Munich and has expanded to three locations worldwide, including Vancouver and Los Angeles. Scanline VFX is well known for its proprietary software that enables creation of lifelike fluid effects, and the company was nominated for an Academy of Motion Pictures award in 2008 for its work on Clint Eastwood’s “Hereafter.”

The process of producing and editing visual effects involves a massive amount of digital content and data. With large numbers of renders and simulations constantly taking place, as well as the need to expand their workforce on a moment’s notice, Scanline VFX requires a secure, redundant and scalable infrastructure. To support the needs of their clients and create better workflow, Scanline VFX sought to establish one centralized location for all their servers and equipment, and make their network and systems accessible to their worldwide employee base.

Security

Creating visual effects for high-profile movie and TV projects requires tight security for systems and networks. Scanline VFX needed to ensure a secure environment while also providing controlled access for its more than 400 employees to remote into the server without compromising the safety of the network. Scanline VFX can rely on remote hands management from Internap for round-the-clock assistance with access control.

Redundancy

With all their employees relying on access to the server to do their jobs, power blips or outages are not an option for Scanline VFX. To prevent lost productivity, the company set up render farms in Internap’s Los Angeles data center, a fully redundant facility. Data centers designed with concurrent maintainability eliminate single points of failure, and ensure that if a component goes offline, a backup component will still be able to function and prevent service interruptions. As a result, Scanline VFX’s environment remains up and running even in the event of a power outage. In addition, the company is backing up petabytes of data to Internap data centers as an extra step to ensure business continuity.

Scalability

As Scanline VFX takes on more projects, it needs the ability to add more employees and increase capacity on a moment’s notice. With large volumes of digital content and high CPU usage, the process of scaling infrastructure requires more power than a traditional colocation footprint. Data centers designed for high power densities can provide additional power as business needs grow in the future. In addition to scalable density, Scanline VFX also needs access to clean power, because electrical pollution and power surges can have a negative impact on sensitive electrical equipment.

As a result of moving its infrastructure to Internap, Scanline VFX has been able to achieve the security that comes with true redundancy for its environment. The ability to scale allows the company to expand its footprint with little notice to meet the demands of its clients. “We’re in the business of making VFX for films; we’re not in the business of managing our data centers and dealing with AC requests and power issues and creating redundancies,” said Scott Miller, Studio Manager for Scanline VFX.

At the OpenStack Summit in Vancouver this week, we heard several announcements about the promising future of OpenStack. In case you missed the Summit, or didn’t get a chance to catch up on the news, here is a collection of articles to keep you updated.

OpenStack Launches Certification Program And Identity Management To Ensure Interoperability

One challenge for OpenStack has been providing clarity regarding how different cloud computing solutions work together. This week, OpenStack announced two initiatives that address the interoperability issue. First, there will be a series of interoperability tests to ensure core elements of OpenStack from different vendors will interoperate. In addition, a new distributed identity offering will enable users to create instances on other clouds using their “home” OpenStack cloud login.
Read entire article here.

OpenStack gets bare metal, storage support

The OpenStack Summit in Vancouver kicked off with a list of initiatives aimed at extending cloud services capabilities based on the open-source cloud platform. Internet infrastructure services provider Internap unveiled a new bare-metal cloud service on the OpenStack platform to deliver high-end Infrastructure-as-a-Service (IaaS) options. Bare-metal cloud infrastructure is touted as giving DevOps teams an on-demand IaaS option for OpenStack, where they can deploy and manage cloud-native and enterprise applications in hybrid cloud environments.
Read entire article here.

OpenStack simplifies setup with Community App Catalog

To increase OpenStack adoption, the OpenStack Foundation announced the OpenStack Community App Catalog, which will be a repository of common applications and app-stack components used when building a cloud. OpenStack COO Mark Collier says the catalog will allow users to share best practices, while also helping new users get started. This project may show prospective users that the utility of openstack can make it easier to stand up applications and systems.
Read entire article here.

OpenStack isn’t just ready for enterprise adoption, it’s already there

While deploying OpenStack cloud isn’t easy, there are several businesses that are actively using OpenStack for large-volume, mission-critical workloads. With so many vendors committed to OpenStack, deployment will become easier, and the concerns voiced by Forrester and others are being addressed by the OpenStack community. The challenge of porting legacy systems to a new platform will be a problem regardless of whether you choose OpenStack or another cloud provider such as AWS.
Read entire article here.

OpenStack Embraces Containers

The message from OpenStack COO Mark Collier in his keynote presentation was to think of OpenStack as an agnostic integration engine. Demonstrating the ability to
While interest in containers has been seen as a threat to OpenStack, that is not the case, based on what we’ve learned at the Summit. Ultimately, OpenStack could offer developers a single API to manage their workloads, regardless of whether they run on containers, virtual machines or a combination.
Read entire article here.

As bare-metal cloud gains momentum within the OpenStack community, Internap is proud to announce the extension of our bare-metal offerings to the OpenStack platform.

With bare-metal AgileSERVERs from Internap, devops teams can provision and manage bare-metal instances, along with Internap’s AgileCLOUD public cloud instances, through the OpenStack Horizon management dashboard. With up to 10 VLANs, AgileSERVERs offer easier scalability for complete network control, without having to deploy custom infrastructure. And, with hourly billing options, you don’t have to worry about contract terms.

Performance

OpenStack-based bare-metal AgileSERVERs offer developers another IaaS option for building flexible, high-performance environments. Based on observations from our customers as well as validation from third parties, dedicated bare-metal infrastructure provides high levels of consistent performance, which can benefit data-intensive and latency-sensitive workloads. This is particularly advantageous for scenarios where transactions must take place within a limited time window, such as big data and adtech use cases.

Flexibility and choice

Extending the benefits of bare metal to OpenStack gives devops teams more flexibility to achieve the required performance for their apps. With dedicated bare-metal server instances, developers can build more customized cloud environments that support a wide range of applications with different needs.

Hybridization

Furthering our commitment to increase OpenStack adoption, our OpenStack-based cloud portfolio – AgileCLOUD and AgileSERVER – will support OpenStack Kilo by the end of this year, giving you even greater scale and hybridization capabilities across the entire OpenStack ecosystem. We believe that a hybrid approach is an efficient way to build a best-fit infrastructure and look forward to the advantages that interoperability and federated identity support will bring to the OpenStack platform.

Internap has been a pioneer of the bare-metal space, and our developers are among the top 40 contributors to the OpenStack code base with more than 400 commits to date. Providing high-performance infrastructure is our top priority, and we’re excited to bring together our expertise on bare metal and OpenStack.

Data center cooling is one of the biggest challenges of high-density data center design. The approach to cooling depends on your requirements, and in a high-density data center, this includes the kilowatt (kW) per rack that you intend to support across the facility.

Let’s discuss some data center cooling challenges and considerations for high-density environments.

Stranded capacity

This refers to lack of flexibility in the relationship between space, cooling and power distribution within a data center. In other words, space, power or cooling is available but unusable, due to one of the other items being unavailable. As an example, chillers may have remaining cooling capacity, but the amount of air handlers deployed is not enough to meet the requirements of the IT equipment.

Power distribution units (PDUs) may have remaining electrical capacity, but remote power panels (RPPs) are out of breaker positions, caused by underutilized 3-phase circuits as an example. A flexible design is imperative to make sure a data center can support the changing needs of customers and avoid stranded capacity.

Hot aisle containment

As power densities increase above 6kW, hot aisle containment becomes a better option to keep hot and cold air from mixing. Temperature control in high-density environments becomes increasingly important to ensure efficient performance of servers and other IT equipment, and a fully contained hot aisle will offer more tolerable working conditions throughout the data center floor.

In the event of a cooling system failure, this approach can also provide a slightly longer window of time before temperatures become intolerable.

Data center zoning

Many data centers are mixed environments designed to accommodate different power densities. Data center zoning refers to the practice of designating certain sections of the data center floor for higher power densities. These sections are designed to offer more power per cabinet and provide the appropriate cooling for the density required.

To address this challenge, one method of zoning is to include under floor walls or baffles to maintain the required cooling for that zone within the area of the raised floor that it needs to cool. This technique is complimented with a hot aisle containment to return the warm air to the air handlers that are also providing the cooling.

The challenge with any high-density environment is providing the appropriate temperature and air volume at the inlet of the servers while rejecting the hot air away from the ambient air of the room. If managed properly, this approach will provide effective equipment operation and tolerable conditions for data center and IT technicians.

Within the next few years, the Internet of Things (IoT) could have implications across several areas of our personal and professional lives. To keep you in the know on this topic, we’ve compiled some relevant articles that offer different perspectives on the impact of this rising technology.

What is the Internet of Things?

Need an “Internet of Things 101”? What is it, and why does it matter? This article explores how the IoT can potentially benefit your life on a day-to-day basis, as well as some of the challenges that will be encountered as the technology evolves. The realities of IoT will affect not only our homes and appliances, but businesses as well, including industries such as manufacturing and healthcare. Read entire article here.

A New Frontier for CRM: The Internet of Things

According to Gartner, Customer Relationship Management (CRM) technology will be an integral aspect of supporting the emerging IoT technologies. The IoT marks a paradigm shift that will affect how customers interact with brands. The advent of connected technology will require more sophisticated methods of serving customers. CRM solutions will help brands manage customers through multiple sales and service channels. Read entire article here.

First Click: you can’t spell ‘idiot’ without IoT (Internet of Things)

Are all these new connected devices worth the cost? As an alternative perspective on the IoT hype, this new technology offers to fix something that isn’t really broken with an expensive, complex solution. Is the race to put a chip in nearly every household device really making our lives better? Read entire article here.

Economist: The Internet of Things will deliver surge of productivity

In this article, Harvard economist Michael Porter discusses how the IoT can reignite our economy and encourage innovation. IoT data and predictive analytics will fuel efficient product development based on real market drivers and needs, ultimately reducing waste within the economy. This will result in opportunities for growth and productivity. Real entire article here.

As an early leader in the bare-metal cloud space, we at Internap are pleased to be one of only a handful of providers profiled in Forrester’s new research report, Consider Bare Metal as a Viable Cloud Option.

As more applications and workloads demand high-performance infrastructure, the market for bare-metal cloud has increased rapidly. More vendors are claiming to offer bare-metal capabilities, and like the common practice of “cloud washing,” infrastructure is now being “metal-washed.”

Not every provider who advertises bare-metal cloud can deliver it. Consumers should be wary of vendors that are metal washing other services such as dedicated hosting, which, unlike bare-metal cloud, is often sold on a term contract and offers zero automation or elasticity.

Internap is proud to be part of the small but notable class of bare-metal providers covered in Forrester’s report.

Virtualization does not equal cloud

As we experience the evolution of dedicated servers, cloud is no longer synonymous with virtualization. Bare-metal cloud isn’t virtualized, and offers the ability to deploy dedicated bare-metal servers in cloud-like fashion, giving access to single tenant physical resources without noisy neighbors or resource contention. Available on an hourly basis, bare-metal cloud delivers the automation and elasticity of traditional virtualized cloud, along with the higher IOPS and disk I/O required by performance-sensitive workloads.

For certain segments of our Internap customer base, high performance is crucial. Adtech, online gaming and big data workloads rely on dedicated infrastructure with extremely low latency to make sure their transactions take place within a certain time limit. Forrester refers to these as “bounded transactions,” and for some of our customers, this window of time is 75 milliseconds. Bare-metal cloud offers significant advantages regarding performance and cost efficiency for these types of transactions.

As an early leader in the bare-metal cloud space, we believe Forrester’s report serves as validation of what we’ve observed across our customer base for the past four years. Bare-metal cloud is also gaining momentum within the global OpenStack community.

High power density data center design will become increasingly important as technology continues to advance. As we mark the 50th anniversary of Moore’s Law in 2015, the number of devices connected to the Internet shows no signs of slowing down, and colocation facilities must be prepared to support the growing demand. Here at Internap, we’ve seen a significant increase in power draw across our data centers, and we even have customers that are exceeding 20kW per cabinet.

Advantages of high power density

Colocation customers that choose a high-density environment typically have a lower total cost of ownership and see greater returns over time. Additionally, as their business expands and requires more servers and power, customers can scale their infrastructure without having to invest in a larger colocation footprint.

High-density design challenges

Establishing a high-density data center requires careful planning. Design considerations are critical to ensure proper cooling, electrical distribution and redundancies. Let’s take a look at some design challenges that should be considered when planning for high densities.

Cooling. Faster processors generate more heat, which makes cooling and temperature control extremely important. High-density data centers require more cooling in a smaller footprint than traditional facilities. Initial costs for cooling equipment will likely be more expensive, since special cabinets may be required to handle higher temperatures.

Plan ahead for cold or hot aisle containment, which will be necessary when densities get closer to 20kW/rack. Design considerations should include the flexibility to add chimneys at each cabinet to remove the hot air upward through a return plenum in the ceiling.

Response time. In the event of a cooling failure, temperatures in high-density environments will rise rapidly. For example, if the chiller pump or CRAH fan fails, the data center operator will have only minutes before 20kW/cabinet regions start to exceed ASHRAE recommended temperatures. For this reason, consider feeding CRAH fans and chiller pumps from a separate infrastructure UPS, or develop a means to introduce outside air through fans on battery or generator back up power.

Power distribution. High-density data centers require a flexible distribution design to alleviate and prevent electrical issues. Be sure your team knows how to manage circuits and determine possible overload situations prior to installation. By planning electrical architecture together with rack layout, your team can better manage conduit and circuit locations. In addition, the route of conduit and piping should be considered so that underfloor obstruction is limited.

As power densities increase, so does the risk of inadvertently overloading the infrastructure, which makes careful planning even more vital for high-density environments. The considerations described here are only the beginning.