The data center industry is evolving as a result of new technologies and advancements. Increased storage capacity, market expansion, green data centers, hybrid IT and the rising demand for cloud services are all shaping the future of the industry.

Below is a collection of articles to help you stay informed on news and trends affecting the data center industry.

Room to grow: Tips for data center capacity planning
Learn several factors to consider when planning your data center capacity needs in this book excerpt, The Practice of Cloud System Administration: Designing and Operating Large Distributed Systems Vol 2. A clear understanding of your current usage, along with the specific aspects that should be tracked, can help avoid unpleasant surprises. This helpful excerpt also provides a formula you can use to calculate the capacity needs for your resources, along with important industry terms you should know.

Data center wars – Battleground Houston
The data center market in Houston is heating up, and several firms have invested in additional infrastructure to keep up with the high demand. More than half of Houston-based companies still maintain in-house data center operations, and the increased competition could mean better rates and more competitive colocation services.

Watch this video to learn more about green data center design
Designing a high-performance, sustainable data center requires a team of experts and careful planning.
Sustainability consultants work with the design team early in the process, to provide expertise regarding energy, material selection and site selection. These decisions will ensure that the completed project meets the standards of specific certifications, including Green Globes^®, LEED and ENERGY STAR.

How to find hybrid harmony in the data center
In this edition of Wired Innovation Insights, Jelle Frank van der Zwet discusses the importance of hybrid environments in the transition to cloud platforms. Data centers designed to support hybrid IT will serve as a catalyst for enterprises to connect on/off-premise legacy infrastructure with public and/or private clouds. The “multi-cloud future” will require hybrid IT capabilities so that organizations can realize the benefits of cloud solutions.

Cloud starts taking bite out of data center construction
What is the future of data center construction, and how will the shift to cloud services impact the industry? According to a recent IDC survey, fewer companies will run their own infrastructure by 2018, which will increase the need for cloud services. While data center construction will decline for enterprises, cloud service providers will likely need to expand their facilities to meet the higher levels of demand.

Learn more about the future of colocation by downloading the white paper, Next-Generation Colocation Drives Operational Efficiencies.

Earlier this month, we launched our new Managed DNS service. DNS (Domain Name System) is used to convert easily remembered URLs like “internap.com” into IP addresses. Anyone that has an online presence uses DNS. Unfortunately, DNS is often overlooked and can become a hidden source of latency for websites and Internet applications. With many web applications becoming more complex, they increasingly pull data from numerous domains. Too many DNS queries can cause bottlenecks and delay users that are trying to access your content.

Broad footprint and global anycast

To address these issues, our Managed DNS is supported by 24 points of presence worldwide and uses global anycast. Global anycast ensures that your users get directed to the closest name server (rather than broadly distributing requests), thereby improving DNS performance and uptime. If your website or applications reference a number of other domains or you have a user base that spans across many regions or continents, our Managed DNS service can help your users get the best browsing experience possible.

Configure your DNS directly from our customer portal and/or use our API to significantly reduce time spent integrating Managed DNS with your hosting environment and web applications.

Key benefits of Managed DNS include:

• Fully managed network with Managed Internet Route Optimizer^TM (MIRO) minimizes latency and improves DNS consistency.
• 24 POPs worldwide so that DNS queries are sent to the closest name server.
• Support for multiple record types including A, AAAA, NS, MX, CNAME, TXT and SRV.
• Intuitive web interface to allow you to create, publish or modify zone files and record types to map your domains.
• Customizable API provides the control and flexibility you need to quickly get your websites up and running.

Internap is happy to introduce two new instances for its OpenStack-powered public cloud, AgileCLOUD. With increased CPU, RAM and storage capabilities, these new instances offer customers enhanced choice to better meet performance and scalability demands.

As part of the Series A flavors, the new A1.16 instance comes configured with 16vCPU, 16GB RAM and 320GB of ephemeral SSD storage. Instances such as this are ideal for small to medium-sized databases and websites that require moderate network throughput.

For more demanding workloads requiring additional memory and higher network throughput, the new B1.16 instance is a better fit. It comes configured with 16 vCPUs with dedicated threads, 60GB RAM and 320GB of ephemeral SSD storage. Like the other Series B flavors, the new B1.16 instance also offers SSD-based persistent storage for specific storage needs.

Global Availability

The recent expansion of Internap’s OpenStack-powered public cloud footprint to Amsterdam enables A1.16 and B1.16 instances to be immediately available in Europe as well as Internap’s Dallas facility. With capability to also manage OpenStack cloud through an integrated Horizon dashboard, Internap now delivers greater flexibility and advanced cloud management features to all developers and enterprise customers.

We recently discussed the future of colocation during a webinar with a panel of industry experts. Guests included Internap’s own Randy Ortiz, Philbert Shih from Structure Research, Mark Hurley from Schneider Electric and Sean Brady from Cushman & Wakefield.

We experienced one problem during this webinar – the panelists were so engaged in conversation that we ran out of time to answer questions from the audience!

Below is a question-and-answer session with Randy Ortiz, VP of data center design and engineering at Internap, to address some of your inquiries from the webinar.

Q: How do local utility companies support data center power requests?
A: The first step in a data center power request is to first understand how much power you will need when your data center is completely built out. For example, if a data center is designed for 10,000 square feet and the maximum IT design load calls for 120watts per square foot, then the ultimate design IT load will be 1.2MW and an additional 1.9MW using a 1.5 PUE to calculate the power needed for cooling. Add another 100kW for miscellaneous loads and you will need approximately 2.2MW of power for your data center. A much more detailed calculation and letter will be required by your engineer to submit to the utility. Once the utility receives and approves the power request for this capacity, the utility will work with your design team to provide the proper feeders, electrical equipment and transformers for your site.

Q: Is having multiple electrical substations an absolute requirement for a Tier 3 data center?
A: According to the Uptime Institute’s definition of a Tier 3 facility, multiple utility feeders are not required for a Tier 3 or a Tier 4 site. What’s most important is having the required number of generators to support your site as prescribed for each Tier level for utility backup.

Q: Is a designed Tier 3 data center okay, as opposed to an actual certified Tier 3 data center?
A: In my opinion, you can validate a Tier 3 facility by requesting drawings and a basis of design from the data center provider that can then be validated by any professional licensed engineer.

Q: What is your perspective on the emergence of colo providers offering “shared services”? Is this a pathway for more value-add colo services?
A: I like the idea of providing as many IT services as the customer needs to help them have a turnkey data center solution. Service providers will continue to help differentiate themselves and add more to their bottom line by offering services that customers demand. If there is a market for it, then colocation providers will offer it.

Colocation is only one piece of the puzzle. Smart organizations are planning ahead for cloud or managed hosting services that they may need in the future. These services are a good complement to colocation, and hybrid hosting capabilities can offer long-term cost savings.

The OpenStack Summit in Paris last week was an exciting time for Internap. AgileCLOUD, our Openstack-based cloud platform, launched in Amsterdam, expanding our global reach to meet growing demand for public cloud services. Our London-based team had the opportunity to showcase the new Horizon cloud management portal, available natively through AgileCLOUD. With so many exciting announcements during the Summit, here is a collection of our media highlights.

Internap Rolls Out European Cloud Offering and OpenStack Control Panel
What better place than Paris to announce the European expansion of AgileCLOUD? The OpenStack-powered cloud is already available in Dallas and Montreal, and the new Amsterdam location marks the first facility on the other side of the Atlantic.

Announcements Flow as the OpenStack Summit Begins in Paris
Among other announcements, Internap, which provides Internet infrastructure services, has revealed that it is expanding its OpenStack-powered AgileCLOUD footprint to Amsterdam. This marks the third location for AgileCLOUD worldwide.

Internap Expands OpenStack Cloud Platform to Amsterdam, Integrates Horizon Dashboard
Horizon is the official OpenStack management dashboard, and Internap says it is one of a few OpenStack cloud providers to expose the native console.

Internap Rolling Out Cloud Node in Amsterdam
Philbert Shih from Structure Research offers insights on the launch of Internap’s new node in Amsterdam, Netherlands for its AgileCloud product. This geographical diversity should come in handy for both data location and performance reasons.

Cloud Bytes: ADVODA, TI Sparkle, Internap, Midokura, ON.Lab, vXchnge
Internap is taking its cloud capabilities to Europe. They are expanding their OpenStack-powered AgileCLOUD into Amsterdam while adding a new integraged Horizon dashboard. That includes both the virtual cloud servers and those bare metal ones, offering greater flexibility of infrastructure.

Over the past decade, communication technologies have given rise to a wide range of online services for both individuals and organizations via the Internet and other interconnected networks. Network routing protocols play a critical role in these networks to effectively move traffic from any source to any destination.

The core routing decisions on the Internet are made by the Border Gateway Protocol (BGP) as specified in IETF RFC 4271, which uses routing tables to determine reachability among autonomous systems and make route selections. BGP guarantees that the traffic will go through the shortest path to reach its destination; however, it does not guarantee that the route is optimal in terms of performance (e.g. latency, loss, etc.) and/or costs as shown in the following figure.

Internap’s Managed Internet Route Optimizer^TM (MIRO) was specifically designed to overcome this problem by evaluating different path characteristics to create performance metrics that are used to select the best routes for Internap customers.

What is MIRO?

MIRO is a highly engineered, distributed system whose functionality can be separated into four core subsystems: Route Collection and Injection, Traffic Estimation, Performance Measurement and Route Optimization. The following is a greatly summarized description of each subsystem:

Route Collection and Injection

MIRO actively learns full BGP tables (prefixes) announced by each provider to be aware of the different routes available to each destination. There are different ways to learn this information including direct BGP sessions with edge routers or via SNMP queries. Also, this subsystem is in charge of updating routes (moving routes) by telling the routers which provider is preferred for each route.

Traffic Estimation 

To estimate the volume of traffic, MIRO consumes network flow information from the edge routers (e.g. Cisco Netflow, IPFIX). The flow information contains source and destination IP addresses, port numbers, octets, etc. This information is aggregated into subnetworks (prefixes) that should match the ones collected by the Route Collection subsystem, and then the total amount of traffic to each destination is calculated and handed over to the Route Optimization Engine.

Performance Measurement

Performance metrics can be defined as a combination of one or more measurement variables like latency, packet loss, jitter, etc. MIRO selects target IPs on each destination network for which it collect performance metrics, and does so via different techniques including pings and traces. This information is then combined and normalized, and handed to the Route Optimization Engine.

Route Optimization 

The Route Optimization Engine is the brains of the MIRO system. It consumes routes and provider information, traffic estimates, performance metrics, and user rules and parameters, and runs a mathematical model to find the absolute best route for each destination. The Route Optimization Engine then sends the selected routes for the destinations to the Route Injection subsystem, which makes sure the changes are applied.

In order to optimize each component and meet quality and performance requirements, our engineering efforts had to overcome many challenges, including:

• How to accurately calculate traffic at the prefix level, a problem which is still an open issue in the academic and research community;
• How to optimize routes in polynomial time considering there are hundreds of millions of possible solutions;
• How to keep track of thousands of route changes per minute from several providers without negatively impacting our edge routers’ performance; and
• How to calculate convergence points for target selection to ensure stable and reliable probing for collecting performance measurements.

One of the main differences with its predecessor is the way new MIRO optimizes routes. Our previous method, a heuristic TCP/IP route management control, worked with BGP in an automated manner. It updated routing tables with the best performing routes available to provide a superior alternative to the manual route selection approach that many data centers employ to compensate for BGP’s inherent deficiencies. The new method is a deterministic approach based on a mathematical model, expressed with a linear programming formulation that considers performance, cost and efficiency as required.

With the completed deployment of this new MIRO system in all our markets, we immediately confirmed a better performance and much faster response time to network events. In Atlanta (ACS) for example, we selected a random day to show the best case average latency for all carriers compared against MIRO, and as expected, MIRO had better performance from 2 up to 15 milliseconds faster:

In the previous figure, you can see there is a network event on provider RED (represented with the red line), where the average latency increased from 120 milliseconds to approximately 180 milliseconds. If we look at the amount of traffic MIRO was putting on provider RED, we can notice it reacted to the event almost instantly, moving about 2.8 Gigabits of traffic per second to other providers (from 4.3Gbps to 1.5Gbps).

Similarly, we selected the same day in New York (NYM) to compare the average packet loss per provider against MIRO. MIRO’s average packet loss is 0.01% versus 0.04% for the rest of the providers:

MIRO brings our customers faster and more stable gaming networks, Content Delivery Networks (CDNs), social networks and general availability for end users as a result of consistent low latency and packet loss. Even though the Internet wasn’t designed for speed, MIRO addresses the deficiencies in BGP and routes traffic along the best available paths.

We’re excited to announce the immediate availability of the Horizon cloud management portal for our Next-Generation AgileCLOUD nodes in North America and Europe. This new portal provides an additional set of tools to manage cloud infrastructure with OpenStack-specific features. OpenStack power users will appreciate the ease of use and flexibility provided by Horizon. Those who rely on the OpenStack API for specific features will be excited to have an OpenStack-native interface to complement our AgileCLOUD portal.

In our initial launch of Horizon, it will provide native access to several unique OpenStack features described below. As Next-Generation AgileCLOUD evolves, the Horizon dashboard will provide access to more features until they are ported into our mainstream customer portal.

1. Security groups

Security groups will enable simple role-based layer 3 and layer 4 firewall management rule sets that can be applied to one or more instances. Instead of managing firewall permissions on each server instance, security groups can be set with predefined firewall rules and then can be applied to groups of servers as needed.

2. SSH key management

If your organization practices SSH Key Management for access to infrastructure, you’ll want to start using it with Next-Generation AgileCLOUD right away. As you launch new instances, you can immediately apply SSH public keys (created within OpenStack or uploaded from your existing keys) so that you can provide immediate access to authorized users.

3. Configurable VLAN management for server groups

Internap has always provided public and private VLANs for AgileCLOUD customers. Now you have more flexible configuration options, by creating instances that should only be used on a private network (such as a Database server). Combine this with security groups to provide extra access control over your infrastructure.

4. Advanced volume management

Many of our customers have asked for more advanced volume management. Of course you can still choose between a local ephemeral storage disk and a persistent root volume, but now you can create additional persistence block storage disks and attach them to your server instances. Use this to provide additional data storage, or use software RAID to stripe/mirror your disk volumes.

5. Instance resizing

Instance resizing allows you to resize your server instance to a new/different configuration. For example, if you start with a small configuration but later find you need to increase the CPU and memory, resizing it will allow you to apply the new resources in minutes.