November 2021 - databank-next.go-vip.net

Anxieties Keep IT Up at Night

A great application is like a top quarterback in the NFL. The performance stats will be off the charts—but only if there’s a solid supporting infrastructure, i.e., talented blockers on the offensive line.

Just like head coaches with top quarterbacks, IT teams with great applications often worry about performance, availability, and costs. These anxieties are often created by a misalignment between applications and their supporting infrastructures. Finding the right balance to keep costs under control while making sure applications meet end-user needs often proves elusive.

The First Step to Align Applications and Infrastructures

If you’re feeling overwhelmed by the number of application issues you need to address, you’re not alone. The first step in aligning your workloads to the right infrastructure is to examine your application. The right data center partner can help you work through the key questions so you can find the answers you need:

What is the role of the application workload in your business?
Is the workload mission-critical to business operations?
How mature is the application within the development lifecycle?
What challenges are faced by your application—cost, performance, scalability, security, compliance?
What are the disaster recovery requirements?
Does the application integrate with other applications running in different infrastructures?
Is the application portable to another infrastructure environment
Who manages and maintains the application?

If you already considered these questions and are leaning toward a particular environment, be sure to examine and validate your reasons. The benefits that a particular environment delivers today may not hold up over the long term.

For example, if you already operate your infrastructure on a public cloud platform and use an auto-scaling service like Kubernetes, consider your peak requirements for CPU, RAM, and storage IOPS. You can then check with a private cloud provider to see what they charge to handle those requirements. If you use steady-configured resources, a colocation data center provider can advise which resources would function cost-effectively in a private cloud or on dedicated bare-metal servers.

Prioritizing Cost vs. Performance Factors

By answering the key application questions, you can assess the priority of cost control vs. the ability to quickly scale up and down as spikes in activity occur and then dissipate. You can also evaluate the balance of the cost against user-response times and the amount of downtime you can afford.

The maturity of your application is another factor. Early in the lifecycle, when first building an application, IT determines the infrastructure costs to support the minimum viable product of the application. If the application is built in the cloud, the cost will be low at first because there isn’t any demand for the application, and it requires a small amount of compute resources.

Then, when the application matures and the demand rises, you can model the costs over a longer time period. This gives you a better idea of which components belong in a data center and which components belong in the cloud.

Also Consider Your Infrastructure Management Capabilities

In addition to asking questions about your application, consider how you manage your infrastructure environments. Do you have internal resources to handle the task? Are you already working with a managed service partner, or are you looking for a new partner?

The infrastructures you select for your application may be influenced by who has access to manage the environments and if they can do so effectively. Sometimes you have the flexibility to manage an environment yourself; in other cases, you are limited to working with the hosting provider, whether it’s a public cloud, private cloud, or colocation data center. Finding a partner that gives you multiple options enables you to more easily create a hybrid IT environment, which is usually the best option.

For additional information on aligning applications to the right infrastructure, check out our white paper, How to Align Your Applications to the Optimal Infrastructure Environments. By adjusting your paradigm and asking the right questions, who knows? You just might give your quarterback…that is, your application…the supporting infrastructure it needs for off-the-charts performance.

To capitalize on 5G technologies and deliver low-latency experiences for customers, software-as-a-service and content providers need the ability to provision services from the edge as close as possible to their customers.

Raul Martynek, CEO at DataBank, continues to refine the definition of edge computing. This is the third of a three-part series looking at the edge from the multiple perspectives of enterprises that use and drive edge development: hyperscalers, network providers, SaaS firms, and content providers. In this article, we examine how SaaS and content providers can leverage the low-latency edge to fully capitalize on the capabilities of 5G networks.

Get Close to the Edge to Capitalize on 5G

As the World Economic Forum and PwC emphasized in a jointly-published white paper, 5G enables unprecedented levels of connectivity. This includes superfast broadband, ultra-reliable low-latency applications, massive machine-type communications, high reliability, high availability, and efficient energy usage. Together, these defining features will transform many sectors, such as manufacturing, transportation, public services, gaming, healthcare, and financial services.

With these capabilities, 5G also promises to greatly expand the data processing functionality and interactive capabilities that software-as-a-service (SaaS) and content providers can deliver to their customers. A key factor in enabling this transformation is the ability to provision services as close to the edge as possible when necessary.

Workload demands will spike as dense customer clusters try to access applications and content. Just think of how many Apple TV subscribers wanted to watch Ted Lasso every Friday as new episodes became available!

Edge Compute and Storage Resources Reduce Latency

With 5G bringing low latency and greater bandwidth to wireless customers, the technology is also paving the way for greater consumption of SaaS and content services in markets with sub-par internet service. Many people may actually move away from high-population areas, simply because 5G networks will support their bandwidth needs.

SaaS and content providers should thus prepare to serve the needs of customers moving to rural areas. But even where 5G is available, user experiences could degrade if services are not provisioned close enough to the edge. By adding customer-facing compute and storage resources to colocation and modular data centers operating at the edge, SaaS and content providers can improve customer experiences through improved application performance.

Rounding Out the Edge Perspectives

Our first article in this Voices of the Industry series covered the perspective of hyperscale cloud providers. We discussed how the definition of edge computing is expanding to include localized hyperscale data centers. This change comes as the IT industry once again turns toward decentralization to deliver applications as close as possible to end-users.

In our second article, the edge from the perspective of network providers, we discussed how network service providers want to move away from traditional, centralized compute infrastructures. Instead, they are embracing a more specialized, decentralized model.

As we now examine the edge from the perspective of SaaS and content providers—including those that deliver gaming, entertainment, and augmented reality services—we can see that they are looking for some of the same attributes as the hyperscale cloud providers. SaaS firms and content providers also want a wide footprint in Tier 1 and Tier 2 metros to get near where most of the US population lives. And like the carrier network provides, SaaS firms and content providers also value secondary colocation data centers in carrier hotels as well as private cloud options that offer on-ramp connections to public clouds.

Provisioning Services As Close As Possible to Customers

To fully capitalize on the benefits of 5G technologies that reduce latency and offer advanced caching capabilities, SaaS and content providers are mostly focused on the edge that puts their workloads even closer to their audiences. This requires highly geo-specific infrastructures such as modular data centers that can be placed anywhere—on the roof of high-rise buildings, at the base of cell towers, and within cable headend facilities.

Given the propensity of customer clusters to suddenly expand and contract, SaaS and content providers also need the flexibility to move workloads across secure and compliant hybrid environments that feature containerized infrastructures. No one knows exactly where future customer demand will come from, and the public cloud isn’t always going to be the answer.

A hybrid environment addresses this challenge by providing access to a mix of colocation, modular, and on-premises data centers as well as private and public cloud infrastructures. Such an infrastructure environment also facilitates managing and moving workloads to the ideal environment as they evolve across the application lifecycle—from development to testing, staging, and production.

Key Hybrid Infrastructure Attributes

For SaaS firms and content providers looking to deploy their applications within a hybrid infrastructure to get close to the edge and take advantage of 5G networks, it’s best to collaborate with an infrastructure partner offering a range of IT environments that complement public cloud deployments:

Colocation data centers covering the entire geographic area of the customer base
Multi-tenant and private cloud environments to adjust for security requirements
On-ramp public cloud interconnects to accelerate file and data exchanges
Modular data centers to reduce application latency for clustered and rural users

Portability across environments is another vital attribute. In some cases, public cloud vendors entice customers to code applications to sync with the provider’s proprietary platform. This means the application can’t be migrated to another environment without extensive recoding.

The maturity of an application is also a factor. When first building applications, IT allocates the infrastructure costs to support the minimum viable product. If the application is built in the cloud, the cost will be low at first because there isn’t any demand for the application. When the application matures and the demand rises, the company can then model the costs over a long time period to get a better idea of which components belong in a data center and which components belong in the cloud.

Bringing Applications and Content to Customer Doorsteps

5G along with edge infrastructures that provision applications close to customers will have a major impact on the services SaaS and content firms can deliver. This is critical because customers will demand faster and faster responses and expect more and more interactivity from their applications. The combination of 5G and edge solves this challenge by unlocking new capabilities for applications like augmented reality as well as faster data crunching and greater workflow-processing capacity.

Gone are the days when networks could send customers all the way to data centers on the other coast and then back again to their home office. 5G and low-latency edge networks virtually put applications and content right on customer doorsteps. And that opens the door to much greater customer experiences.

This article was written by Raul Martynek, CEO, DataBank. Mr. Martynek joined DataBank in June of 2017 as the CEO and is a 20+ year veteran in the telecom and Internet infrastructure sector, having held senior positions at several communications and networking companies, as well as asset management firms. Raul earned a BA in Political Science from Binghamton University and received a master’s degree in International Affairs from Columbia University.

Frequently Asked Questions

Security Is Only as Strong as the Weakest Link

The Scottish philosopher Thomas Reid once said, “A chain is only as strong as its weakest link.” So too is your ecosystem of vendors.

Every company faces many IT risks and challenges when sharing digital assets and integrating systems with third parties. If vendors let their guard down when exchanging data with your IT systems or providing software and hardware devices, cybercriminals could very well find their way into your IT infrastructure. The hackers can breach one of your vendor’s digital assets first, traverse their connection to your organization, and then hide in your network.

A good example of supply chain risk is SolarWinds, which provided an IT security tool called Orion used by 33,000 customers, including the U.S. Government. In December 2020, Russian hackers compromised the Orion infrastructure, which resulted in numerous IT organizations and security service providers suddenly questioning the toolsets on which they were relying. This highlights the importance of understanding supply chain IT security.

Multiple Levels to Secure

The make-up of your IT supply chain will vary depending on your industry and your business model. In essence, the chain includes any vendor that provides you with data, an IT asset, or online services. This includes data exchanges and application interfaces as well as the hardware components in your servers, desktops, tablets, smartphones, and other devices.

They all present opportunities for cybercriminals to breach a system, which, in turn, may allow those with ill intent to attack one of your systems. It is, thus, critical to ensure every entity in your downward supply chain meets your security requirements.

Validating each of their security postures and their compliance with pertinent regulations is crucial; not all vendors have sufficient budget and resources to secure every IT asset properly. Each vendor also has its own set of downstream suppliers, representing your fourth-party and fifth-party supply chain. They can all potentially have weak links that threaten your IT infrastructure.

For example, a SaaS company engages a managed services or colocation data center provider (third party) to host enterprise applications provisioned to customers. The data center provider also relies on a vendor (fourth party) from which they buy server hardware to run the applications. That server vendor depends on another vendor (fifth party) to provide antivirus software. If there’s a security weakness at any point in this supply chain, it can threaten the infrastructure of every other company across the supply chain.

Validate Vendors Though Independent Auditors

As you work with your vendors to verify their security posture, be sure to carefully construct your SLAs so vendors know what you require for a security posture, how security should be maintained, and that you need to verify the posture on a recurring basis. Also, check if your vendors follow a security framework such as NIST, ISO, or CSA STAR (See acronym glossary below). If not, that’s a red flag their security capabilities will require close examination. Therefore, ensure they can prove compliance to one of these standards via a third-party audit and that you can gain access to that report.

Following a framework like NIST tells you a vendor has good cyber hygiene. Another burgeoning, robust framework is CMMC, which is also based on NIST standards. It’s a third-party risk management framework with which the federal government requires all Department of Defense (DoD) contractors to comply. Like NIST, CMMC can also be leveraged by the private sector.

To validate a vendor’s security posture, determine if an independent auditor has attested to the correct design and deployment of the controls. Going one step further, some organizations provide an extra layer of assurance by validating auditors.

For example, SSAE auditors are monitored by AICPA while A2LA verifies FedRAMP and CMMC auditors, also known as 3PAOs (third-party assessment organizations). Reports generated by either of these groups give you additional peace of mind—you know the auditor is an experienced professional who is not just rubber-stamping what they heard from the vendor.

Maintaining a Strong Link Within Our Customer Supply Chains

At DataBank, we base our security methodology on NIST 800-53. Following this framework, we apply 325 security controls across all our colocation and managed services data centers. Many of those controls also roll down to our supply chain partners, depending on the type of service or hardware they provide. The controls can include any of the eighteen NIST categories such as access control, physical and environmental security, security awareness and training, and the application of other specified security controls.

We validate our vendors by examining their SSAE auditor reports. We also make our SSAE and CSA STAR audit reports available to our customers, as well as proof of compliance with PCI-DSS and HIPAA. These audit reports make it easy for customers to evaluate our security posture within their supply chains.

Getting Supply Chain Ecosystems to Work Together

Third-party risk can be tricky because many parties want to blame someone else when risk is exposed. Eventually, supply chain partners need to realize they are part of a complex ecosystem that needs to get on the same page to protect each other.

Such teamwork is vital because we all face a constant cat-and-mouse game with cybercriminals, defending against the many known and unknown vulnerabilities that keep emerging. The good guys will keep battling the bad guys, and to keep up, we need to rely on frameworks and multiple tactics to ensure our downstream and upstream digital assets stay secure.