Why Ad Tech loves zcolo
How edge Markets are On The Rise
zColo's latest facility in Waco may be just the location many risk managers have been waiting for.
7 Exciting Trends in Iaas
What changes you can expect to see in Infrastructure-as-a-Service in the new year
2016 / Volume 1 / Issue 1
Finding colo space in Germany
PUBLISHER & MANAGING DIRECTOR
Send comments and inquiries to:
zColo: The Magazine is published twice a year by the
zColo Product Management & Marketing teams.
Copyright 2016 by Zayo Group. All rights reserved.
Consequently, rather than build their own data centers in these smaller markets, many content providers want to partner with colocation providers, like zColo that are in the right geographic locations, with the necessary density and can offer connectivity, scalability and reliability, as well as square footage and power. To learn about Zayo’s colocation options in Tier-1 and Tier-2 markets, check out our zColo facilities list.
How More Companies are Pushing Their Content to the Edge
Exploring the trend of colocation in edge markets
COVER: Why Waco?
Marcus Anderson on his vision of our latest Texas facility
7 Exciting Trends in Infrastructure-As-A-Service
What to be on the look for
zColo Facility in France Leads the Way in Energy Efficiency
Besançon just got greener
How zColo Can Help Your Company Prepare for Any Disaster
What's the Importance of Germany to the Telecoms Industry
Frankfurt & Dusseldorf change the look of telecom
Cloud Compute vs Cloud Storage
What's the Difference and Why Should It Matter to You?
Speed, Scalability & Savings
How Ad Tech Gains a Competitive Edge with zColo
Expanding our Presence in Chicago
Learn more about 840 S Canal St, Chicago
Ask A Data Center Manager
Meet Joshua Fernandez in Denver
Ask A Data Center Manager
Thank you for reading the first issue of zColo: The Magazine. zColo is introducing innovative solutions around the world, and this magazine is a means for letting you, our customers and partners, know about our exciting growth.
It’s been a busy year! We added three new U.S. facilities in 2016 in a quest to keep up with demand:
In the U.S. we’ve focused on developing our portfolio around three unique types of data centers –
Downtown interconnect-focused facilities
Suburban, scale-out facilities
New “midtown” options that offer high scale power and low latency networking.
All of these facilities, in conjunction with the Zayo network, are enabling the development of powerful new options for our clients.
In addition to new facilities in the U.S., in 2016 we also expanded into five additional countries:
Finally from an organizational perspective, Zayo’s Cloud division welcomed a new business leader, Lisa Neal-Graves. Under Lisa, our Cloud business has continued to flourish and grow. From object-based-storage to fully customized private cloud deployments, our Cloud team is meeting the needs of a global marketplace with excitement and energy. Today, Zayo boasts cloud nodes in nine markets across North America and Europe.
We’re excited to release this issue and hope that you will find it interesting and informative.
Thank you for your continued business,
Senior Vice President, zColo
fROM MY dESK
How More companies are Pushing their Content to the Edge
Driven by bandwidth-intensive services and streaming, Internet traffic has continued on a dramatic growth trajectory. According to a recent Cisco report, global IP traffic is expected to pass the zettabyte threshold by the end of this year and should reach two zettabytes by 2019.
As more people adopt streaming and real-time entertainment in their personal lives, and rely on connectivity, file sharing and big data in their professional lives, businesses are upgrading their communications infrastructure to provide fast, high quality services.
In addition to high capacity connectivity, businesses are increasingly adopting colocation and cloud services, which provide greater capacity at less cost than traditional, in-house hardware and software platforms. When it comes to data centers, many businesses are adding capacity in both traditional Tier 1 markets and, increasingly, to Tier 2 markets – growing population centers like Nashville, Minneapolis and Denver.
Zayo founded its data center business, zColo, on a highly interconnect-oriented model of colocation. As we have evolved organically and through acquisition, we have added a unique mix of facilities to our portfolio. We now have 62 highly connected data centers in the US, Western Europe, and Canada ranging from expansive suburban 2N facilities to resilient network-focused sites in major commercial districts. One of our newest stateside facilities, located in Waco, Texas, complements nicely our growing portfolio in Dallas and carries with it some very interesting attributes.
As a side note, we have invested heavily in Dallas in all forms of
communications infrastructure, including an almost $200M fiber-to-the-tower project. We now have 5 data center facilities in the market, including the former Stream facility at 1100 Empire Central Pl. Our strategy in Dallas is similar to many other major markets where we operate: we provide larger, more robust colocation deployments in the downtown vicinity and tether back to space-constrained or expensive carrier hotels leveraging Zayo’s dark fiber, or 10G/100G wavelength network.
Dallas is the 6th largest colocation market in the world - beating out Los Angeles, Chicago and Singapore. Enterprises and carriers alike are tripping over themselves to get space in this ultra-connected, centrally located market.
So why were we talking about Waco - nearly 100 miles south of Dallas?
Waco is a small city in Central Texas, situated between Dallas, Austin, and Houston. The home of Baylor University, this community has primarily been a college town for over a century. What I didn’t realize is this town’s unique geographic location makes it connectivity’s best-kept secret.
Many important networks travel straight through Waco, including Zayo, AT&T, FiberLight, Time Warner Cable, and Level3 - just to name a few. This makes Waco an important junction for networks traveling in and out of the Dallas metro area.
However, beyond the clear connectivity benefits, it still wasn’t clear to me why we selected this facility in particular. There are lots of markets with connectivity value. What else made Waco such a valuable find? The answer: Disaster Recovery.
In 2008, Hurricane Ike made landfall in Galveston (less than an hour southeast of Houston). In fact, since 1980 at least 70 tropical or subtropical cyclones have affected the state of Texas. Hurricane Ike was blamed for at least 195 deaths,113 in Texas alone. Damage from Ike to the US Coastline totaled nearly $30 billion. It was the third costliest Atlantic hurricane of all time.
Waco has a high level of natural protection from hurricanes, as it sits outside of the the 500-year flood zone. Our data center, located at 700 Austin Avenue, only has a 0.02 percent annual chance of a flood. The area is also well protected by levees.
Surrounding metro areas seem to always have power issues. Recently, heavy storms knocked out power for over 20,000 users in Austin. Additionally, more than half a million users lost power around Dallas when two thunderstorms hit North Texas within four days of each other.
Waco operates on a power grid separate from Texas’ major city systems. This provides an advantage during potential blackouts or brownouts. The facility incorporates a dual transformer commercial power feed configuration to provide extra resilience to critical power systems. The commercial power is backed by dual diesel generators with 2000KW 3-phase electrical ratings to provide uninterrupted power during any potential power outages or brownouts.
But What if disaster does strike waco?
Waco has been no stranger to natural disasters. In May 1953, a deadly tornado struck Waco, destroying 2,000 buildings and causing 114 deaths. To mitigate future disasters, our facility was built to withstand the worst the world could throw at it. And at the time, that meant natural disasters and nuclear warfare.
Because it was purpose built as a nuclear fallout shelter, the facility can withstand a direct hit from a Category 4 tornado. It has steel reinforced concrete construction and pillars that run from the roof of the building through the basement floor and an additional 33 feet (10 meters) below the building to rest on the bedrock beneath along with four foot skirts at the top. The additional skirt at the top allows for weight distribution on the floor and anti-pancaking of the floors above in case of catastrophic damage to the building. The floors of the facility are constructed on 18 inches (45 cm) of steel reinforced concrete.
zColo’s Waco facility lies outside the 500-year flood zone, sits on its own power grid, can withstand a tornado and nuclear war. On top of all this it sits on the communication highway connecting Dallas to the rest of Texas. When considering these benefits, the more appropriate question may be: “What took us so long?”
Since adding 700 Austin Ave to our portfolio, this is often the facility I find myself discussing the most. Customers everywhere ask me what our best facility for disaster recovery is and for me it’s a simple discussion. When they inevitably ask, “Why Waco”, I can’t wait to tell them. Waco is not Toronto or New York City or Paris; nonetheless, we’re very enthusiastic about this new addition to our portfolio of data centers.
Click Here to request a tour of our Waco facility.
In large part, it’s to improve quality of high-bandwidth Internet services to users outside of the top metro areas. Providers are caching their content and services in these Tier 2 data centers, which are extending the “edge” of the Internet further from the Tier 1 hubs of New York, Silicon Valley, Chicago, Dallas and Miami. This approach improves speed and latency, requirements for high def streaming, fast twitch gaming and a host of cloud-based services for businesses.
Service providers of all stripes, along with enterprise organizations—particularly small and mid-sized companies—are pushing the Internet expansion across a broader geographical base. According to a recent report by Ovum, the global analyst firm, the enterprise migration to the cloud is not the only factor affecting data centers and the associated demand for colocation services. “The explosive growth of video is…fundamentally changing how networks are constructed.”
Over the past 2 years, zColo has added seven facilities in Tier-2 markets in the U.S., including:
3 in the Denver metro area
2 in Orange County, California
1 in Minneapolis
1 in Waco, Texas
zColo’s footprint now includes nearly 40 data centers in the U.S. and another 20 in Europe and Canada. From carrier hotels in the heart of major connectivity hubs to highly-secure disaster recovery campuses, all of our strategic locations offer a variety of solutions for any user.
Consequently, rather than build their own data centers in these smaller markets, many content providers want to partner with colocation providers, like zColo that are in the right geographic locations, with the necessary density and can offer connectivity, scalability and reliability, as well as square footage and power. To learn about Zayo’s colocation options in Tier-1 and Tier-2 markets, check out our zColo facilities list.
Director of zColo Product Management
Zayo's Texas Fiber
7 Exciting Trends in
Infrastructure as a Service (IaaS)
IT infrastructure has undergone a dramatic makeover in the past decade, as bleeding edge infrastructure as a service (IaaS) solutions ratchet up the speed, power and capacity of enterprise IT. Now that we’re entering the second half of the decade, it’s hard to imagine how dramatically the business technology landscape will change by 2020. A number of IaaS trends are making us excited about the possibilities.
From technologies that enable greater availability and performance to greater adoption of the hybrid cloud, many exciting things are happening in enterprise infrastructure as a service. Here are seven IaaS trends we’re watching right now:
The evolution of hybrid cloud solutions: By combining public and private cloud services, colocation and managed hosting, hybrid cloud models offer the flexibility, scalability and performance today’s businesses demand. That’s why Gartner predicts that half of all enterprises will adopt a hybrid infrastructure as a service solution by 2017.
Companies that may have held off on updating their IT infrastructure due to the performance and security limits of the public cloud — or the scalability limits of the private cloud — are seeing the possibilities of hybrid IT and hybrid cloud infrastructures, which right-size business workloads to the IaaS environments they best fit. This creates a best-of-all-possible-worlds scenario for businesses that run a mix of persistent, bursty and compliance-driven workloads.
The possibilities of the multicloud: By allowing businesses to take advantage of cloud hosting across multiple data centers, the multicloud builds redundancy and resilience into infrastructure as a service while mitigating latency due to geographic location. Providers that perfect a centralized strategy to streamline the management and security of multiple cloud hosting environments will be the big winners of 2017.
The cloud takes up Big Data: While “Big Data” was the buzzword du jour only a few years ago, it’s rapidly becoming mainstream. Infrastructure as a service has moved to accommodate today’s Big Data demands, with cloud solutions providing the compute, network and storage resources needed to make data analytics available and accessible on demand.
BYOD steps aside for liquid computing: The consumerization of IT is continuing its occupation of the workplace. As more CIOs grapple with BYOD (Bring Your Own Device) initiatives, liquid computing — that is, seamlessly accessing data and applications across multiple devices — is further blurring the line between consumer and enterprise technology. What’s the takeaway for Iaas?
Storage takes a major step up: Along with Big Data analytics, business is demanding greater performance and capacity from infrastructure as a service. Autonomic storage tiering is making it happen. By analyzing data, placing it on the appropriate storage tier within the same volume and enabling the migration of data from tier to tier, autonomic storage tiering (aka hybrid storage) improves the performance and flexibility of your storage infrastructure.
Business continuity and shrinking CapEx budgets drive IaaS growth: A recent study by Infiniti Research predicts growth of infrastructure as a service to reach as much as 42 percent over the next five years. What’s motivating this growth? It’s the benefits achieved by converting capital expenditures (CapEx) to operating expenses (OpEx), combined with a focus on business continuity.
By building redundancy into enterprise IT and combining hosting strategies, today’s hybrid infrastructure as a service solutions can help streamline your business continuity and disaster recovery program. And by helping to reduce your data center footprint with high density data centers, cloud hosting, private cloud infrastructure and managed hosting, IaaS reduces the amount of CapEx you need to invest into on-premises hardware and data center space, converting more budget to OpEx.
Technology is enabling better security: One of the biggest IaaS trends we’ve noticed in our business is that more customers are adding managed IT security services to their infrastructure as a service solutions. While high-profile data breaches (most recently in the healthcare vertical) cyber attacks are alerting businesses to the importance of high-quality data security, outsourcing is helping them bolster their security profiles.
Kermit was right. It’s not easy being green.
Tough as it may be, we know that sustainability is incredibly important both for the environment and for maximizing operational efficiency. That’s especially true for our 62 zColo data centers, where we make continuous improvement. Our focus includes efficient power system components, environmentally friendly building materials that reduce overall power loads and free cooling that reduce or eliminate compressor chilling in colder seasons and increase efficiency during summer months.
We’re proud of our first advanced green data center. zColo’s Besançon facility in eastern France introduces state-of-the-art cooling, significantly reducing the power usage effectiveness (PUE), meeting stringent safety standards and optimizing energy efficiency.
This data center uses the Ecobreeze modular cooling solution from our partner Schneider Electric and has a PUE of 1.1. The EcoBreeze unit is designed to live outside the data center and can automatically switch between two forms of economized cooling.
Air-to-air exchange occurs when modules bring in air to the indirect evaporative cooler, then return it to the data center. However, if this form of cooling cannot be done, the air is cooled through indirect evaporative heat exchange. Evaporating water on the outside of the heat exchange channels removes heat from the air. The EcoBreeze unit in Besançon maximizes local climate conditions throughout the year to cool the facility, eliminates the environmental concerns associated with water cooling and requires no chemical water treatment.
PUTTING THE ‘COOL’ IN COOLING
According to a study from the Natural Resources Defense Council (NRDC), we’re on the right track with our approach. From an overall industry perspective, there’s an opportunity to improve energy usage by 40 percent, which could save upwards of $3.8 billion in 2016 alone. Cooling systems are the biggest culprits in terms of inefficiency.
Energy efficiency cooling systems are important not only to curb runaway power consumption but also to accommodate more computing capacity. New forms of cooling will improve the PUE significantly. We’ll monitor the results in Besançon and evaluate adopting this cooling technology in other centers across North America and Europe.
zColo Data Center in France Leads the Way
in Energy Efficiency
Figure 1: States Affected by Superstorm Sandy
States affected by Superstorm Sandy
the Importance of Disaster Recovery
An estimated 30% of companies in states affected by Superstorm Sandy had a data center that suffered a power outage.
In 2012 disaster struck. Not just in one region, but all over the United States. In one year, the U.S. saw a total of $110 Billion in destruction from 11 weather disasters, including a tornado in downtown Dallas, wildfires that burned more than 9.1 million acres in the West and 10 named tropical storms and four hurricanes.
The biggest of them all, Superstorm Sandy, was the second-costliest hurricane in U.S. history. Sandy was directly responsible for over 150 deaths in the U.S., Canada, and the Caribbean, plus an estimated $75 billion in damages.
Business was almost brought to a halt as the communications capital of the world, New York City, felt the wrath of this enormous storm. An estimated 30% of companies in states affected by Superstorm Sandy had a data center that suffered a power outage.
Of those, 65% lost power for 2 days or more, 36% were voluntarily off the grid for 3 days or more, and 19% waited a week or more for utility power to come back on. The storm took many data centers offline, with flooding and power outages – disrupting communications and business.1
Zayo had a strong plan and had made extensive preparations. Our infrastructure is diverse and designed to failover across our broad footprint. As a result, we were able to keep our network and our zColo data centers operational during Superstorm Sandy. Our experience, and our disaster preparedness plan, has helped our customers improve their own preparation and business continuity plans.
The average cost of a data center outage has risen by 46% over the last five years. In fact, for every minute a data center is down, it costs nearly $9,000.2
And yet, in a study by the Disaster Recovery Preparedness (DRP) Council, 73% of respondent organizations worldwide are not taking adequate steps to protect their data and IT systems.3
It’s generally a matter of time and resources. Developing a disaster recovery plan takes effort and knowing how to handle a disaster takes practice.
Fortunately, colocation providers, like zColo by Zayo, have weathered the storm before, literally, and know what it takes to manage a disaster.
“Superstorm Sandy made landfall on October 29, 2012 near Atlantic City, NJ with winds of 80 mph. A full moon made high tides 20% higher than normal and amplified Sandy’s storm surge. Streets were flooded, trees and power lines knocked down and the city’s famed boardwalk was ripped apart… Seawater surged over Lower Manhattan’s seawalls and highways and into lowlying streets. The water inundated tunnels, subway stations and the electrical system that powers Wall Street and sent hospital patients and tourists scrambling for safety. Skyscrapers swayed and creaked in winds that partially toppled a crane 74 stories above Midtown. A large tanker ship ran aground on the city’s Staten Island.” 4
Prior to Sandy, Zayo laid the groundwork for a successful operational response. Its approach reflected Zayo’s comprehensive disaster preparedness plan – and the importance of selecting a data center provider that is well-prepared for disasters ranging from fiber cuts to hurricanes. Our disaster readiness is based on 4 key factors:
Zayo operates 1,800 route miles of owned and operated New York City metro fiber, with redundant routing options. In addition, all three of Zayo’s zColo facilities in the area have diverse connectivity so there are multiple routes in and out of the data center.
Responsive national operations team:
24/7 monitoring over three geographically diverse Network Command Centers (NCCs), and hundreds of network engineers and technicians ready to respond – with enough fuel and supplies to see them through multiple days.
Detailed and regular customer updates:
Zayo shared regular email updates, which were posted to zayo.com and Twitter, detailing emergent issues and Zayo’s response plans. An open conference bridge was established when storm flooding threatened a generator at one facility.
Accountable executive-level leadership:
Zayo’s disaster preparedness plan included a comprehensive playbook, including a clear chain-of-command and accountability. No time was wasted debating what to do.
As a result, Zayo kept both its fiber network and colocation facilities operational throughout the entire storm. When potential issues did arise, Zayo delivered superior uptime, responsiveness, and transparency. Zayo’s response to an issue at its 111 8th Ave colocation facility demonstrated the effectiveness of this approach and the expertise of its operational team.
Zayo was so proactive in its customer communication that when one of its customers had an emergency need for connectivity during the height of the storm. In less than one day, a contract was requested, quoted, signed, and delivered.
Fiber outages were minimal and only one of Zayo’s colocation customers across its three New York City area facilities experienced downtime.
30% of companies throughout the Northeast had a data center that suffered a power outage as a result of Superstorm Sandy… but not Zayo customers.
Disaster can strike in any number of ways. In fact, the most common disasters include power outages, hardware errors, and old fashioned human error. To truly prepare for the worst, every company needs a detailed DR plan. To keep your company above water, several steps you should take include:
Perform a risk assessment
Test and train
Find a strong colocation partner
Perform a risk assessment - Identify what the internal and external risks are that have the potential to impact your operations.
Figure 2 demonstrates an example risk assessment matrix to help your organization analyze its exposure to disaster.
What are the hazards, their probability, and their magnitude?
Likelihood of events: Scale of 0 to 1 (0 will never occur, 1 will always occur)
Impact of events: Scale of 0 to 1 (0 no impact, 1 complete devastation)
After you have created your matrix, rank your risks and determine the appropriate approach.
Prevent - Work to mitigate
Accept- Maintain vigilance
Contain - Minimize likelihood
Plan - Plan steps to take if occurs
Test and Train
Once disaster management approaches are defined, create a regular schedule for practicing and training your staff, both for new hires and current employees. The more you practice, the more prepared your team will be in the case of an emergency. For example, if your plan involves colocation, make sure your team knows how to contact remote hands for support during a disaster.
Rachel Dines, Writer for CIO.com, writes:
“If you woke up tomorrow and ran a marathon, how would you fare? It’s highly doubtful that you would successfully run the 26.2 miles without months of training, drills, and exercises.
The same is true for disaster recovery (DR): The chance that you could successfully recover IT operations without having exercised your DR plans on a regular basis is slim at best. The chance that you could successfully recover and meet your recovery objectives is zero…”
Here are a few best practices once you have established your DR Plan:
1. Determine specific exercise objectives upfront - Make sure that there are clear and concrete objectives and goals set up front that will help determine the ultimate success of an exercise.
2. Develop specific risk scenarios for your exercises - It is important to define specific risk scenarios even for DR testing for two main reasons: A) It provides a more realistic situation for the response team to react to, and B) different scenarios require different actions from the IT staff.
3. Vary exercise types from technical tests to walk-throughs – Technical tests will tell you how failovers are working, while walk-throughs will tell you how the team’s communication is working.
4. Make sure to test all IT infrastructure concurrently at least once per year - Waiting longer than a year risks too much change in IT environments and personnel; you need to bring new staff members throughout the organization up to speed on DR plans.
5. Identify members for the core DR Response Team - When picking a core response team to lead IT recovery, it’s important to pick people who can work under extreme amounts of pressure. During an exercise or test, identify those individuals who can remain calm and collected.5
Find a strong colocation provider
A colocation provider that has strong track record in minimizing downtime during disasters, like zColo by Zayo, can provide your team with the diverse and secure infrastructure it needs. Consider the following questions when determining if a colocation provider is a strong asset for your DR strategy:
Do they offer a geographically diverse NCC team that operates 24/7?
Do they offer carrier-neutral connectivity?
Do they have a proven track record of maintaining uptimes during disasters?
Do they have a proven track record of leadership engagement during disasters?
How geographically diverse is their data center footprint?
How many facilities do they manage across edge markets, like Austin or Denver?
How have they communicated updates during disasters in the past?
Do they demonstrate creativity and agility when responding to customer needs?
In addition, a good partner can walk your disaster recovery preparedness team through the available options when designing your plan. They will be able to explain how varied locations or technologies, like Disaster Recovery as a Service (DRaaS), can be integrated to reduce downtime and maintain productivity.
Figure 3 demonstrates 3 DRaaS options that can strengthen any DR strategy.
At any moment disaster can strike. Whether a fiber route is accidentally cut, you experience hardware failure, or a hurricane rocks the entire Eastern seaboard, be sure to protect your data and maintain productivity for your entire organization. With the right planning and partners, your team will be prepared for whatever surprises come their way.
1 Building Operating Management. (2013, October). Survey: The Effects Of Hurricane Sandy On Critical Facilities. Retrieved from FacilitiesNet: http://www.facilitiesnet.com/datacenters/article/Survey-The-Effects-Of-
2 Electrical Construction and Maintenance. (2016, January 19). Data Center Outage Costs Continue to Rise. Retrieved from Electrical Construction and Maintenance: http://ecmweb.com/power-quality/data-center-outage-
3 PRNewswire. (2014, March 4). Global Benchmark Study Reveals 73% of Companies are Unprepared for Disaster Recovery. Providence Journal. Retrieved from http://www.providencejournal.com/business/press-releases/20140304-
4 Sharp, T. (2012, November 27). Superstorm Sandy: Facts About the Frankenstorm. Retrieved from Live Science: http://www.livescience.com/24380- hurricane-sandy-status-data.html
5 Dines, R. (2012, January 18) How To Improve Disaster Preparedness. Retrieved from CIO: http://www.cio.com/article/2400373/disaster-recovery/
How zColo can help your company prepare for any disaster
How Do you prepare for the next disaster
Figure 3: Options for Disaster Recovery Using DRaaS
Risk Factor (a*b)
Loss of critical vendor
Restoring Your Critical Data
Data protection provides insurance against things like accidental file deletion, file corruption, virus infection, and more.
Not all hosts are suitable for virtualization.
Storage replication enables the failover of physical hosts such as SQL database servers from a primary site to a secondary site.
Virtualized hosts can be replicated in their entirety, and automated restoration makes fail-over simple and reliable.
Failover testing is non-impactful and can be done monthly, quarterly, or semiannually depending on your needs.
Recovery Point Objective = 24 Hours
Recovery Point Objective = 15 Minutes
Recovery Point Objective = <1 Minute
Recovery Time Objective = Weeks
Recovery Time Objective = Days
Recovery Time Objective = minutes
• Agent based file backups for physical hosts
• Changes are captured and de-duplication minimizes costs
• Customizable backup policies to only back up what’s needed
• Image based backups for virtual servers (VM’s)
• With image based backups, you have a point in time snapshot of the server including the OS image
• Data Retention (Short Term or Long Term)
• Local or Offsite (or both)
• With Storage Replication, data from the primary site is replicated to the secondary/failover site at a point in time
• Snapshots allow for point in time captures of the data at a particular moment, they can be retained for up to 7 days
• Snapshots are often useful when performing maintenance work, to provide a “roll-back” point if something goes wrong
• With Workload Replication, virtualized environments are replicated from one site to another
• As changes happen to the VM’s at the Primary site, the changes are replicated to the Recovery Site
• Workloads can be replicated to public/private clouds
Figure 2: Example Risk Assessment Matrix
2.8 mW total UPC
Generator & UPC Redundancy
SOC 1 Type 2, SOC 2 Type 2 & SOC 3
Remote Hands Services
In Las Vegas nature is not a concern. zColo’s Las Vegas customers don’t worry about earthquakes, hurricanes, blizzards or any other natural disaster.
Our Las Vegas facility connects you to Zayo’s massive network of over 8.4 million miles of fiber. This facility also boasts connectivity to 8 other carriers offering maximum flexibility.
Because of its full name
(Frankfurt am Main) and it being
such a hub of business, locals in Frankfurt will refer to the city as "Mainhattan", and as a nod to its financial status it’s sometimes also referred to as "Bankfurt".
What's the Importance of Germany to the Telecoms Industry?
A little geography
Sometimes the distance between the U.S. and Europe can feel much further than the 3,500-ish miles it is; we want to narrow that gap and bring the two sides of the Atlantic closer.
At around 138,000 square miles - slightly smaller than Montana – and bordered by nine countries sits Germany.
Germany is different to other European countries with an infrastructure that you don’t see in many other geographies; Germany has a range of key cities with their own identities that is partly due to the old east/west divide: Hamburg is a hub for media, Munich for manufacturing, Cologne for gaming, Berlin for politics - and two cities that we’re going to study in more detail - Düsseldorf for telecoms and business and Frankfurt for finance.
Many of the telecom companies based in Düsseldorf started in advertising - along with them there are also about 200 publishers, around 170 national and international financial institutions, 130 insurance agencies and one of the eight German stock exchanges. Düsseldorf is also the hub for the German mobile phone market. Many foreign communication technology companies are based there such as Huawei, NTT, Ericsson, Nokia, and GTS.
Frankfurt is the largest financial centre in mainland Europe; it’s the home to the European Central Bank (the home of the Euro which is one of the largest currency areas in the world), the Frankfurt Stock Exchange and many large commercial banks. The Frankfurt Stock Exchange accounts for more than 90 per cent of the German turnover.
A little history
In January 1998, the German government relaxed its regulations and restrictions around the telecom industry and liberalised the market so private companies could get involved and invest to improve the infrastructure.
Following this, and as a result of subsequent massive investment, the formerly outdated system in the eastern part of the country was swiftly modernized to have the most technologically advanced equipment and infrastructure systems in the world. Zayo’s network infrastructure rollout across Europe began in December 1998 and was completed in March 2000.
A little on the economy
As Europe’s largest and strongest economy it is no surprise that Germany enjoys a well-established telecoms network with its fibre optic cables making up the most dense network in the world; this infrastructure is heavily relied upon to keep Germany’s economy strong and buoyant. To look at this another way, of the world’s largest stock market-listed companies, the Fortune Global 500, 28 of them are headquartered in Germany – for these companies to continue operating at this level, a disrupted or slow telecommunication system is not an option.
It also boasts being the second most populated country in Europe with 81 million people calling the nation home. With all this, it’s clear to see why the German IT and Telecoms market is the world’s fourth-largest scooping 4.3 per cent of the worldwide ICT-turnover in 2014, and that IT providers, telecommunications and Internet services are the second-largest employer in the German industry.
Eighty-six per cent of the German population has access to the Internet – an increase of four per cent from 2010, making it one of the highest percentages in the world. To put this into perspective for comparison 87 per cent of the US population has Internet access, the Spanish are at
76 per cent, and only 18 per cent of Indians have Internet access.
Where we can see a significant upwards trend is the use of data on the go; mobile data has increased from 25 per cent of the German population using it to 64 per cent - we all must just love those cat videos on YouTube on the commute!
Luckily, to keep up with the demand for German-based data and content, Zayo has a carrier-grade European network across Germany and two zColo data centres; Düsseldorf and Mannheim (which sits just outside of Frankfurt).
As OTT (over-the-top) services and especially the financial trading sector grows increasingly important in Germany and companies establishing their main PoP in the city of Frankfurt as a hub, we’re seeing an increase in demand for Germany’s central location in Europe proving important in the wider landscape as we move forward. Thirty-five per cent of German traffic goes through Frankfurt.
If you’re looking at expanding to Germany and across Europe, get in touch with your sales representative today to discover our reach and your options!
Zayo's European Network
Source: World Bank
Internet users as percentage of population
Cloud Compute vs
What’s the Difference and Why Should It Matter to You?
Cloud talk can get kind of, well, cloudy. With so much terminology floating around — from countless “as a Service” acronyms, to consumer cloud services like iCloud® and OneDrive®, to private, public and hybrid clouds — sometimes something as fundamental as the difference between cloud storage and cloud compute can be hard to parse, especially if you aren’t directly involved in the technology on a day-to-day basis.
So what is "the cloud" anyway? The reality is that this nebulous term means different things to different businesses — and the difference between cloud compute and cloud storage is often the most confusing. We'll unpack both concepts and tell you what you need to know to understand how they function both independently and together.
What is Cloud Storage?
Chances are, you’re already familiar with cloud storage, and likely use it every day. The photos you took with your iPhone® on your last vacation, the e-books you’ve purchased from iTunes® and the videos you saved to Dropbox® all live in cloud storage. Typically, when users refer to “the cloud” as it pertains to their personal mobile devices, they’re talking about cloud storage — their files are being stored on a hosting provider’s hardware, sometimes spanning multiple servers or physical sites. This allows your files to be accessible from the Internet regardless of where you are or what device you’re using, while saving space on your own personal device.
Cloud storage providers offer users and companies storage services on a pay-as-you-go basis, with self-service features and on demand scalability. When you need more space, you simply purchase what you need from your provider and instantly access the added storage. For individual consumers, this offers a quantity of space most people don’t have access to on their home computers or smartphones. For businesses, it means cost savings in terms of hardware you don’t have to power, cool, manage or maintain. You can store or archive your company’s data in the cloud for a service fee, and your provider takes care of the administrative and infrastructure costs. An important characteristic, however, is that cloud storage also tends to be “higher volume / lower cost” storage because while data has to be available, retrieval doesn’t have to be instantaneous or fast. It can wait.
What is Cloud Compute?
While you can think of cloud storage as sort of an off-site unit holding, essentially, lots of data — documents, graphics, videos, etc. — cloud compute is an off-site server hosting whole workloads and processes, including business applications and platforms (Software as a Service/SaaS, ERP, analytics. Think applications like Salesforce®.). Now, you’re not just talking about cloud hosting services in terms of space, but also in terms of processing speed, RAM, duration and scale.
With cloud compute, you host workloads like website traffic, enterprise-wide applications like email and CRM or database workloads on a provider’s hardware, and both manage and access those workloads from a variety of locations — for example, from your office or from a remote laptop. Like cloud storage, cloud compute offers pay-as-you go, self-service resource allocation, so you can dial your resources up and down as needed on demand without having to purchase new hardware.
Why the Difference between Cloud Compute and Cloud Storage Matters
In a way, it’s understandable that people commonly confuse cloud computing with cloud storage. For most users, cloud storage is their most frequent—and tangible — touchpoint with the cloud. Further muddying the waters is the way both services work together. Cloud storage is just one important parallel of cloud compute. In addition to the processing power you need to host a workload, you also may need storage for the data associated with that workload. But as we mentioned before, it’s important not to confuse cloud storage—which is accessed via APIs or front-ended with an application (think Dropbox)—with storage which is directly tied to runtime compute resources (typically referred to as “internal storage” or SAN).
With that said, when we talk about “the cloud” solely in terms of storage, we’re missing out on the wide array of capabilities that cloud compute offers especially to businesses. It’s a popular framework to think of the cloud as merely a place to store files—only extended to the day-to-day applications, processes and workloads that keep your organization successful and efficient and enable more advanced services like DRaaS.
With hybrid Infrastructure as a Service (IaaS) solutions combining cloud hosting, colocation and managed hosting, Zayo helps businesses build IT infrastructures with both the cloud storage and cloud compute they need to get the job done.
Contact Zayo today to learn more.
Speed, Scalability & Savings
How Ad Tech Gains a
Competitive Edge with zColo
Progression of Ad Tech infrastructure growth
An Expensive Dilemma
Digital advertising has changed the landscape of available media for marketers everywhere. The days of Don Draper are long behind us. Today’s sophisticated advertisers have more in common with statisticians than ad men. Data now drives every conversation, and critical matrices define growth and success. This rebirth of the modern advertiser is now enabled by new and innovative software applications, which are changing the game.
Advertising technology, or Ad Tech, has been exploding as an industry since 2010. The Advertising and Marketing industry is ranked as the third fastest-growing industry with a 200 percent growth in revenue and employees.1 Ad Tech revenue is set to grow over 300 percent by 2020 -- up from $30 billion in 2015 to $100 billion by 2020. 2
As advertisers become more data-centric and technology-savvy, Ad Tech applications must increase their capacity to deliver meaningful solutions or they will lose swiftly to the competition. These applications must be accessible and functional in an instant. Any delay in moving data or supplying content could lead to missed revenue. Amazon found every 100ms of latency cost them 1 percent in sales. Google found an extra 0.5 second in search page generation time dropped traffic by 20 percent. 3
To execute high-volume data collection and analysis, buy & sell in milliseconds, and deliver ads across multiple platforms, Ad Tech companies are forced to invest heavily in infrastructure -- which can quickly force CapEx investments and complexity to skyrocket. As a result, companies are then left with three major challenges when it comes to executing critical operations:
Control growing costs
Dramatically reduce latency
Support fast growth
To solve these challenges, leaders in Ad Tech are moving toward scalable infrastructure that provides speed, grows with them, and reduces CapEx investment.
Finding the balance between fast, resilient infrastructure and reduced CapEx investment can be tricky. Many companies start off with a simple colo + 1G fiber strategy. They tend to select data centers near their headquarters -- making server maintenance simple. In addition, they purchase connectivity solutions to deliver their applications or content to potential customers around the world. However, as traffic grows, latency may eventually take its toll on user experience, impacting revenue growth.
Over time, Ad Tech firms may supplement their colocation with public cloud services. They tend to use the public cloud initially to manage peak traffic demands, create development environments, and become more scalable. However, it’s not long before operational costs can become unmanageable. Many turn those dollars over to a capital investment, where they own or lease the equipment themselves. They will then stretch their deployments out to two data centers in geographically diverse markets. This also allows them to decrease latency and improve customer experience. 10G network is deployed to connect the facilities and compute and storage systems are distributed.
Eventually, demand in edge markets increases and connectivity-focused colocation deployments are needed around the world. High-volume 100G connectivity is lit and distances are significantly shortened between content and the end user. Latency continues to decrease, by up to 55 percent compared to the previous deployment models. Private cloud is enabled to drive operational efficiency, offer DR resiliency and store petabytes of ever growing data. Additionally, resiliency of the overall infrastructure actually increases. By using a more streamlined colocation model with huge bandwidth, if something were to go down, compute can simply failover to any of the other facilities without missing a beat.
Due to Ad Tech’s high-density computing needs, companies are utilizing 8 to 10 kw/cabinet --twice as much as the average enterprise load, translating to significantly higher costs and complexity to run their applications. On top of this, they need to dedicate operational spend for a geographically diverse staff to manage all these servers. They have to deal with multiple providers of colocation, cloud, and network services around the world and need people skilled enough in every area to design deployments.
What many smart Ad Tech firms have done is found ways to bundle all these services together, add outsourced remote hands staffing, and save up to 34% on the standard cost of their whole deployment. 4
How Ad Tech companies reduce latency and control costs with Zayo
zColo, Zayo’s data center division, offers over 60 carrier-neutral colocation facilities across North America and Europe. Ad Tech companies around the world have leveraged these facilities, as well as Zayo’s expansive network, to deliver creative cost-saving and latency-reducing solutions.
MediaMath, a leading independent programmatic advertising company, leveraged this expansive portfolio to expand their computing power, reduce latency and control their costs. Over nine years, MediaMath’s client base grew to more than 4,000 diverse global companies. Its technology platform analyzes more than 200 billion customer opportunities daily.
“Zayo’s ability to support high-power density loads and our ability to scale were key factors in winning MediaMath’s business,” said Greg Friedman, executive vice president, Colocation and Cloud Infrastructure at Zayo.
MediaMath decided to expand its colocation into zColo’s Oak Brook, IL facility (a suburb of Chicago). They coupled this with Zayo’s expansive network capability, leveraging 10G wavelength connectivity to key peering and interconnection facilities. “The combination of Zayo’s connectivity and colocation was a big differentiator for us. While we are moving our data center to the edge, Zayo’s infrastructure ensures that we have the performance needed to process, move and store massive volumes of real-time data,” said Keith O’Neill, senior engineering manager at MediaMath. 5
OpenX, a global leader in creating programmatic advertising marketplaces, expanded its data center footprint with zColo to support the growing volume of transactions on its OpenX Ad Exchange. The exchange facilitates trillions of transactions yearly for the company’s more than 1,000 publisher clients worldwide.
Given Zayo’s interconnect-rich data center portfolio and extensive fiber network we were well positioned to support OpenX’s growing needs. They decided to expand their dense footprints at zColo facilities in Ashburn, VA and Oak Brook, IL.
Shane Garoutte, vice president, technology operations at OpenX, said, “Zayo continues to be a valuable partner for OpenX. Zayo has proven to be adaptable to the needs of our business while offering a competitive price compared to other providers, making the decision to expand our use of their platform an easy one.”6
Tackling the Holiday Rush
One of zColo’s Ad Tech customers sees a huge traffic spike each year during the holiday season. However, for the rest of the year, network demands are significantly lower. In addition, they wanted to distribute their computing across multiple regions, with a more streamlined design. They found themselves overpaying significantly for bandwidth they weren’t using and infrastructure-heavy colocation facilities they didn’t need. They decided to leverage Zayo’s unique data center footprint and innovative network solutions to control costs and fit their needs.
By moving to four smaller colocation deployments, they were able to distribute their computing power to the edge more efficiently and save money on infrastructure they weren’t using. Since they coupled their colocation with diverse connectivity, if disaster were to strike, their key applications could simply roll over to another facility, rather than everything going down.
The customer also found great value in FlexConnect, Zayo’s commitment-free, usage-based ethernet for cloud and data center connectivity. By leveraging FlexConnect they were able to pay for only the traffic they used each month, rather than a flat fee for bandwidth they may not fully need. This allowed them to ramp up their bandwidth for a few months and easily turn it off when it is not needed. They payed a small fee for ports and just 3 cents per gigabyte for data traffic.
By combining four small colocation deployments around North America and 50G FlexConnect, this customer was better able to manage their costs and the onslaught of holiday traffic they receive every year.
The Future of Ad Tech Infrastructure - and the world at large
As petabytes of data storage quickly turn into exabytes, latency decreases to almost immeasurable speed, and growth in mobile computing continues, enterprises are looking to communications infrastructure to future-proof their operations. The future of colocation and network infrastructure looks to leverage mobile tower technology in new and fascinating ways.
Today colocation comes in three primary flavors:
Large, infrastructure and security-rich facilities in the suburbs
Smaller, interconnection- focused carrier hotels primarily located in the central business district
Midtown facilities that offer the best of both with low latency, low cost and high scalability
Connecting the suburban or midtown facilities to the downtown carrier hotels, called tethering (shown), allows customers to leverage lots of power and redundancy for core computing needs while accessing a plethora of network connectivity options.
In the future large, primary computing data centers will remain, but they will connect to small hubs located near fiber towers. The network will connect directly into the tower itself, pushing content over the mobile network with lightning speed. Dense fiber will allow more content to be delivered faster over wireless networks.
The potential benefit to Ad Tech could be industry-changing. Ninety percent of consumers have stopped using multiple mobile applications due to poor performance7, often related to lag. A reliable, consistent mobile experience is central to a company’s brand reputation and financial performance. By pushing content directly to the tower, mobile network latency could eventually become imperceptible to the end user. Zayo continues to work with its wireless partners to extend and densify their networks, which will improve quality, reliability, speed and performance for years to come.
For every 100ms of latency it could cost emerging Ad Tech applications one percent in sales. But the cost of scaling can be insurmountable. Companies can decrease their latency and control their costs by addressing
all elements of their infrastructure: connectivity, colocation and cloud. Zayo offers all three solutions and has a track record of integrating them to generate the most value to our customers.
1 Graham, K. (2015, September 4). Ad Tech Firms are 2015 Fastest-Growing Companies in the US. Retrieved from MonetizeMore: http://www.monetizemore.com/blog/ad-tech-firms-are-
2 Loechner, T. (2015, April 8). Ad Tech Revenue To Grow To $100 Billion By 2020. Retrieved from Real-Time Daily: http://www.mediapost.com/publications/article/247365/ad-t
3 Liddle, J. (2008, August 27). Amazon found every 100ms of latency cost them 1% in sales. Retrieved from GigaSpaces XAP Blog: http://blog.gigaspaces.com/amazon-found-every-100ms-
4 Based on list rates for individual services versus combined expected TCO for Colo + Network from zColo.
5 Zayo Group. (2016, 23 June). MEDIAMATH SELECTS ZAYO FOR COLOCATION AND CONNECTIVITY. Retrieved from Zayo.com: http://www.zayo.com/news/mediamath-selects-zayo-for-
6 Zayo Group. (2016, March 15). OPENX SELECTS ZAYO FOR DATA CENTER EXPANSION. Retrieved from Zayo.com: http://www.zayo.com/news/openx-selects-zayo-for-data-
7 Whittle, D. (2014, July 16). Nearly 90 Percent Surveyed Stop Using Apps Due to Poor Performance. Retrieved from APM Digest: http://www.apmdigest.com/nearly-90-percent-surveyed-
Fiber to Tower with Colocation Hubs
Zayo will open its fourth data center in Chicago, adding over 23,000 total square feet and two megawatts (MW) of critical power at 840 South Canal Street. The expansion is driven by strong customer demand in Chicago, including a new agreement with a global software analytics company as an anchor tenant in the space.
The new site, which is already on Zayo’s fiber optic network, has low-latency connections to Zayo’s downtown data center at 600 South Federal Street and Chicago’s telecommunications hubs, including 350 East Cermak Road. Inclusive of the expansion, Zayo’s Chicago data center assets grow to an aggregate of 268,000 total square feet and approximately 31 MW of power capacity within the Chicago market, including the existing Federal Street location, as well as suburban facilities in Oak Brook and Mount Prospect. The data centers provide access to Zayo’s diverse, high-count fiber network in Chicago.
Zayo has experienced strong data center demand in Chicago, which is a hub for financial and enterprise infrastructure because of its central location. Demand drivers also include the growth of cloud computing and as-a-service companies for both primary and backup facilities.
“Chicago is a nexus for high-bandwidth traffic peering and connectivity, including our long haul dark fiber, which is driving demand in the region,” said TJ Karklins, senior vice president, zColo at Zayo. “Our customers benefit from extensive interconnections within our data centers, access to our global network and an integrated portfolio of communications infrastructure services.”
Click Here to request a tour of 840 S Canal St, Chicago
Expanding Our Presence in Chicago
Data Center Manager - Denver, Colorado USA
I am a business-focused entrepreneur with a passion for technology and using it to make life easier. Born and raised on a surfboard off the sandy beaches of Hawaii, the snowcapped Rockies have reeled me into colorful Colorado. In my spare time I enjoy being outdoors and experiencing nature; from downhill mountain biking to backcountry snowboarding I love a good adrenaline rush. I'm a big foodie and have grown to love spicy green chile accompanied by one of the many delicious microbrews this state has to offer.
My first steps into IT started in my kindergarten classroom as the appointed computer guy. I took that hobby throughout my education and found ways to make it lucrative. Starting the summer before freshman year I began working with the high school's computer science program as an all-inclusive technician to help with installations and builds, through troubleshooting with students and faculty on daily and complex issues. During my collegiate years at Oregon State University I joined the university's Media Service team as a service technician that provided support to the campus and it's faculty around the clock. Our team provided help desk support at all hours providing remote assistance when available, and hands-on support when necessary. During the out-of-school breaks and seasons our team would provide maintenance to all of the classrooms and handle the needed updates both software and hardware related.
By the end of college I began to realize that my craft offered a valuable skill set that I could take with me into the work force and couple with my educational background in business, specifically entrepreneurship and management. That passion for technology lead me to the brink of my professional career with a network operations team that supported multiple colocation facilities as well as offering managed hosting as a product in multiple markets across the United States. Within this team I gained valuable knowledge on data centers and the process and procedures, along with the compliance needed to operate the facilities. Today, I manage a fantastic team of skilled, zColo professionals that provide network, systems, and operational support to thousands of customers scattered across the globe 24 hours a day, 7 days a week, 365 days a year.
1. How do I add my colleague to my ACL?
A new point of contact (POC) can be added by the Access Controller 1/Access Controller 2 by submitting their POC form to firstname.lastname@example.org. The POC form will specify the new users' access roles and privileges to the customer's account.
Customers can gain a hard copy of the POC form at the local facilities or request a copy through the ticketing system that can then be signed and uploaded. Additionally, a digital copy of the POC form can been filled out and signed through the customer support portal to ease the user management process.
2. How do I get the newest audit report?
All of the audits for each zColo facility can be requested through a customer's account manager, who will pull the most recent copy from our legal team.
Our Denver DEN01 and DEN02 facilities currently hold the following certifications to stay in compliance with our customer's needs: PCI, HIPAA, SOC 1 Type 2, SOC 2 Type II, and SOC 3.
3. How often should I evaluate my power consumption? Can the NOC help with that?
Depending on a customer's growth path power consumption could be a metric needed daily. A customer can request visual inspections of their PDU and live power draws as a remote hands request at any time, and is common for most customers on a monthly to quarterly basis to evaluate their environment.
Customers that neglect power consumption and capacity management within their environment could run into power circuits being over drawn; facing possible loss of redundancy and/or over-utilization charges. To prevent this a simple remote hands request from our data center technicians on a regular basis to observe power usage is recommended. Our remote hands services can be used for activities such as a quick power check to full rack installation and configurations of hardware. The robust skill set of our technicians offers value to our customers taking load off of their
IT requirements in the time of need.
4. What are the best places to eat near the Denver facilities?
DEN02 - "Maggie Smith's Irish Pub" down the street has bread pudding that is to die for.
DEN01 - "Wingin' It" for some scrumptious buffalo hot wings.
Downtown Denver/Champa - "Mercantile" at Union Station for their mind blowing braised short rib sandwich.
Honorable Mention - "Falling Rock Tap House" for a cold pint and the largest selection of microbrews in Denver.
If you have a question for our data center managers that you would like to include in the next issue, email it to email@example.com
ASK A DATA CENTER MANAGER
Las Vegas, NV
Los Angeles, CA
New York, NY
1621 18th St
Denver, CO 80202