White Paper: Cutting through the Fog | NETSCOUT

White Paper: Cutting through the Fog

White Paper: The Need for Real-Time SaaS Visibility and Monitoring | NETSCOUT

To foster greater business innovation and agile response to new business opportunities, enterprises are increasingly deploying Software-as-a-Service (SaaS) applications to meet their needs. This move creates a borderless enterprise, where the traditional line between the enterprise datacenter infrastructure and outside applications blurs. While businesses benefit from SaaS, IT administrators are challenged: who owns the responsibility for SaaS performance and reliability? Enterprises expect IT to step up and own SaaS reliability, especially as it interacts with so many traditional datacenter assets. Currently, IT administrators have little or no insight into SaaS to adequately troubleshoot SaaS problems for their end-users and customers. Proactive, real-time monitoring of SaaS resources with NETSCOUT TruView™ Live bridges the divide between the datacenter and the cloud, giving IT end-to-end control of their networks once again.

    TABLE OF CONTENTS

  • Part I: What's Driving SaaS Adoption?
  • Part II: The Move to a Borderless Enterprise
  • Part III: Time to Take Control of SaaS
  • Part IV: NETSCOUT'S TruView™ Live Solution
  • Part V: NETSCOUT TruView Live Delivers

Part I: What's Driving SaaS Adoption?

Software-as-a-Service (SaaS) is one of the fastest growing technologies in the marketplace. Companies are increasingly moving from the traditional information technology (IT) client- server model of hosting applications in-house to a “borderless enterprise” where software is delivered to them through the cloud from major third-party vendors such as Salesforce.com, Oracle, Microsoft, and dozens of others. And the adoption rate at major corporations is accelerating. According to a recent article in Forbes Magazine, “By 2018, 59% of the total cloud workloads will be Software-as-a-Service (SaaS) workloads, up from 41% in 2013.”1 Total SaaS sales are predicted to top $106B by 2016 - a 21% increase over 2015.2.2

"By 2018, 59% of the total cloud workloads will be Software-as-a-Service (SaaS) workloads, up from 41% in 2013."

Source: Forbes, 2015

The chief reasons for the explosion in the SaaS market are the diversity and innovation of third-party SaaS applications. Initially focused on sales and accounting software, SaaS now encompasses a broad array of cloud-based, enterprise applications, including customer relationship management (CRM), office productivity, mobile applications, and marketing enablement, such as content management (see Figure 1). But why are more enterprises de-emphasizing traditional datacenter IT models in favor of cloud-based applications?

SaaS Quickly Delivers Innovation

The pace of global competition continues to accelerate, and enterprise IT centers are hard-pressed to deploy, configure, and maintain new applications, on premises, in a timely manner. CIOs are caught between maintaining their current IT infrastructure while business development demands greater application flexibility, features, and agility. Adoption of SaaS applications is a quicker, more cost-effective means to a profitable end. The latest in mobile, business process, and productivity applications are just a SaaS menu click away.

With SaaS, enterprises can also easily extend application access to its growing mobile workforce—and most importantly—its customers. Anytime-anywhere access is now a requirement, not a luxury. Sixty percent of all Internet use is mobile.3 Over one billion workers, worldwide, will be mobile in 2015—over a third of the overall workforce. [1.3 Billion Workers to Go Mobile by 2015—Business News Daily]4 Using SaaS, enterprises can more easily extend their application-reach to mobile customers and employees. As cloud-based applications are updated, expanded, or revised, there is no need for mobile users to reconfigure or change their devices. The issue of securely backing up and protecting mobile worker data completely disappears—into a secure SaaS cloud.

Figure 1. SaaS application diversity in the enterprise.

Traditional IT is also struggling to keep pace with end-user demand. Adding, moving, and changing users under a datacenter client-server model can be time-consuming and fraught with issues such as server provisioning, software licensing, backup and recovery of datacenter assets, and ongoing network maintenance. In a recent Harvard Business Review Analytic Services survey report, IT executives cited the following top reasons for using cloud-based computing:

  • “Increased business agility
  • Flexible capacity
  • Faster adoption of new technology
  • Lower fixed costs
  • Lower upfront costs to develop/deploy IT systems
  • Always on newest versions of software without IT updates/patches”
  • Source: Harvard Business Review Analytics Services.5

Clearly, user demand, lowering costs, and greater agility are driving cloud and SaaS adoption. The pay-as-you-go model of SaaS is not just more economical, but adds quick, flexible deployment on an as-needed basis. When employee and partner access patterns shift, SaaS can respond as no traditional client-server architecture can. For these reasons, SaaS is making its greatest in-roads in specific businesses.

Who is using SaaS?

While at first glance it may seem that SaaS capabilities would be a perfect fit for most small to medium-sized businesses (SMB), larger global corporations and enterprises are on the forefront of SaaS adoption. One report shows a SaaS adoption by enterprises at 50%.6 Meanwhile, the adoption rate of traditional mission- critical software in the enterprise is declining. The widely distributed geographical nature of large enterprises, coupled with the on-demand capabilities of SaaS, makes cloud-based services especially attractive to larger organizations. For example, a leading global accounting and corporation can utilize its unified SaaS applications from work teams spread across the world with anytime-anywhere access, rather than invest in on-site software licensing and maintenance at multiple locations. Conversely, small to medium-sized business typically are more geo-centric and have already invested heavily in on-premises infrastructure that they are reluctant to replace or relinquish. However, as SMB infrastructure ages and needs replacement, even those businesses may increasingly turn to SaaS as a viable way to augment their datacenters.

Shrinking market for on-premises deployment of mission-critical software in the enterprise

Figure 2. Source: Data Knowledge Center

Part II: The Move to a Borderless Enterprise

With greater adoption of SaaS by large businesses, they are operating in a borderless enterprise, where the lines between the traditional datacenter and third-party applications have become blurred. And there is no turning back. The borderless enterprise will continue to expand its reach. Enterprises have proven that SaaS fosters better customer engagement and more easily accommodates its growing number of mobile workers. But increasingly sophisticated and innovative uses of the cloud and commercial SaaS applications do not entirely remove burdens on IT. In fact, while the older roles of provisioning and maintaining datacenters may lessen to a degree under greater use of SaaS, this shift presents many new challenges to the IT department.

Cloud, SaaS, mobile, business innovation, and agility are all “must haves” today. But regardless of the origination of the technology, it’s still the business of IT to monitor, maintain, and improve these technologies even if they fall outside the conventional borders of the datacenter. The reason is that virtually all of these borderless technologies impact IT in some way. IT administrators can no longer say that SaaS and the cloud are “not their network, not their problem.” Eventually all activity and data funnel down to central IT—both literally and figuratively (see Figure 3). As a result, IT is burdened with issues concerning bandwidth and wide area network (WAN) providers, telecommunications, insuring and guaranteeing user access, and acceptable performance. When dealing with SaaS and the cloud, IT now must address issues that are “over the border” yet affect the core of the datacenter and its users.

Figure 3. The New Borderless Enterprise: The network has left the building.

Downtime is not an option

End-users and customers do not care if their access originates from a SaaS application or a central server. Performance, reliability, and availability are all that matters. Any degradation in SaaS performance falls squarely in the lap of an enterprise’s IT department. Downtime is not an option. Poor performance means a significant loss in productivity—or loss of customers. Referring SaaS users to the provider will not solve any of these issues in a timely manner, and responsiveness directly correlates with revenue and customer satisfaction.

Although SaaS purchases may have originated outside of IT, business units expect IT to fulfill their traditional role of keeping systems running. For IT professionals, however, the lines are blurred or even erased, making troubleshooting and resolution difficult, at best. Application performance is affected by so many variables that it is difficult to determine quickly where application performance problems originate.

Where is the Problem?

In the complex interaction of users, multiple datacenters, WANs, remote users, and SaaS applications, IT administrators have their work cut out for them. Outages and lagging performance cause a flood of calls to the help desk—but there are no standardized procedures to troubleshoot SaaS problems. Therefore, in the event of an outage or degradation in performance, IT must quickly examine and check the following:

  • The datacenter? When application performance suffers, the first stop for troubleshooting is the datacenter itself. In the case of database-dependent applications, are the logs showing errors? On the network, is the application server responding? Looking at the datacenter network, are routers properly configured or is the switch fabric overloaded? Is the firewall or edge (WAN) switch misconfigured, blocking, or not properly prioritizing traffic?
  • Remote users and offices? The next question is whether the remote network itself is experiencing performance problems. Here, areas to investigate are similar to those in the datacenter: misconfigured firewalls, overloaded network, or in some cases, wireless interference. Are its WAN connections optimized?
  • WAN provider? Has the provider properly configured its provider edge switch to deliver the right bandwidth and quality of service/class of service (QoS/CoS)? Is load balancing in the WAN service functioning? Are segments of the WAN down and rerouting traffic, causing delays?
  • SaaS provider? Of course IT troubleshooters also need to check the status of the SaaS provider. What performance levels does it report? Unfortunately, most service level agreement (SLA) performance checks at SaaS sites only provide general statistics that are not specific enough to properly ascertain how a particular group of users are performing.

So Who Owns the Problem?

The IT department owns the problem. Because of the complex interactions between the datacenter, remote offices, WAN providers, and the SaaS cloud, multiple factors may be contributing to outages or poor performance. Like it or not, performance, reliability, and ultimately the end-user experience are all the responsibility of IT. The last item on the troubleshooting checklist—the SaaS provider—is a blind spot for IT.

IT is SaaS Blind

Given the lack of granular reporting from most SaaS applications, and in some cases WAN telecommunications providers, the IT department is, in effect, “SaaS blind.” With no visibility into the providers’ clouds, there is no opportunity to fix the problem quickly. The general reporting statistics from the providers may show satisfactory levels of service—although aggregated and not specific enough for the IT administrator to pin down their particular problem. This can lead to another round of internal troubleshooting, double-checking visible systems…all the while productivity has ground to a halt. After this exhaustive process of elimination, IT may surmise that, indeed, there is something wrong in the SaaS cloud. But administrators have no way of pinning the specific problem down before picking up the phone to contact SaaS support.

Moreover, Web monitoring tools are also ineffective: they typically run generic testing from the provider to the end-user site. This does not take into account the IT center’s configuration or real usage. These passive tools do not provide active, end-to-end real-time visibility or monitoring.

Lack of Visibility is not a Valid Response

In these cases, the IT administration is caught in a vise. Traditional network monitoring tools are useless once the traffic hits the “fog of the SaaS cloud.” Yet the business holds IT accountable. Regardless of the origin of the SaaS outage or poor performance, lack of visibility is not a valid response. Employee productivity, customer satisfaction, and the operation of the entire enterprise are at stake. For IT to succeed, they need new tools.

Part III: Time to Take Control of SaaS

Part of the architectural problem with SaaS monitoring is not a bug, it’s a feature. Without tight security, SaaS providers would open up their businesses to potential vulnerabilities and threats. Yet that’s no excuse for obfuscating granular performance metrics from its customers. IT administrators need new, real-time tools that give them visibility into their SaaS environment without sacrificing security. In essence, they need to cut through the fog surrounding SaaS to:

  • Ensure network performance
  • Assess service provider performance
  • Deliver efficient, flexible performance monitoring

Ensure Network Performance

Whether onsite or in the cloud, new tools are needed to continually benchmark and monitor performance from end-to-end. When a group of users or even an individual experiences an outage or lagging application response, a single tool should be available to trace and track traffic from one end to the other. Real-time monitoring breaks the cycle of checking, re- checking, and indecision of where problems originate. This minimizes downtime and streamlines IT troubleshooting. IT administrators can then quickly determine whether the problem is inside its walls or if the SaaS provider needs to take decisive action.

Assess Service Provider Performance

Businesses pay for SLAs. Whether it’s a WAN telecommunications provider or SaaS application platform, due to these SLA contracts, businesses have the right to know if they are receiving the agreed-upon level of service. Voice over IP (VoIP), SaaS applications, or even WAN-to-remote LAN connections are bought and paid for on the basis of vendors meeting these SLA obligations.

Real-time monitoring can not only trace the root of productivity-killing problems, but also gives businesses real-time insight into their provider’s true level of service. In many cases, performance degrades gradually, with few complaints from end-users. If Quality-of-Service or Class-of-Service (QoS or CoS) is not performing up to agreed levels, network administrators can put the vendor on notice to fix the problem before it causes a bigger IT headache. By tracking and recording performance levels from vendors over a period of time, businesses may find that the vendor is not meeting its SLA. Businesses can then identify another vendor, or negotiate a better service rate based on unmet SLAs. Alternatively, businesses can work more closely with the vendor to alleviate poor performance when the exact nature of the problem is clearly identified.

Efficient, Flexible Performance Monitoring

With the constant change in the borderless enterprise, any new tools for monitoring need to be just as flexible and agile as the applications that are driving greater innovation. They should be elastic and easy to deploy (and even redeploy) as network conditions change. Key chokepoints, such as end-user access, switches, and WAN traffic should be monitored and tracked from a single pane of glass. Administrators then have a clear and unobstructed view of the entire pathway of network connections.

Part IV: NETSCOUT'S TruView™ Live Solution

When SaaS systems go down, IT administrators are expected to react immediately, putting their reputations at risk. But administrators can’t fix what they don’t see. That’s why NETSCOUT, with decades of network monitoring experience, developed TruView Live: to restore control and visibility to IT. With TruView Live, IT administrators now have end-to-end, real-time monitoring of their entire network-from end-users, to the cloud, and back. TruView Live uses active, real-time network pulses to gather information automatically-or manually-to get to the root of business application performance issues.

With TruView Live, IT administrators can now:

  • Monitor end-to-end availability and performance of networks in real time with a single dashboard
  • Receive alerts and set performance thresholds to resolve issues before they become a user issue
  • Isolate problems and proactively respond to keep the business productive
  • Produce accurate reporting of SaaS performance and SLAs to management, as needed

The end result is that TruView Live takes IT out of reactive mode and can head off many SaaS and network problems before they become a business issue. If a catastrophic outage does occur, IT administrators can quickly use TruView Live to isolate the issue and move rapidly to a resolution, rather than spending hours attempting to track down the causes.

TruView Live and its system of active network pulse technologies are delivered in three distinct architecturally-specific deployments (see Figure 4):

  • TruView Pulse
  • Virtual Pulse
  • Global Pulse
Figure 4. TruView Live delivers real-time reporting from network pulses locally, regionally, and globally.

TruView Pulse

TruView Pulse is a small, active-test micro-appliance deployed into virtually any segment of the network with its plug-and-play 10Mb/100Mb/1Gb Ethernet port. With its agentless architecture, system administrators can plug TruView Pulse into wiring closets, subnets, or anywhere in the datacenter itself. It automatically syncs up and reports to the central TruView Live Dashboard, where administrators can configure each deployed TruView Pulse appliance. With its power over Ethernet (PoE) design, its deployment is extremely versatile, and can be placed and moved wherever IT needs performance data-especially as networks evolve, are reconfigured, or where suspected trouble-spots emerge.

Virtual Pulse

Virtual Pulse is a software-based testing tool that can be immediately downloaded and installed on Red Hat Enterprise, Ubuntu, Amazon AMI, Windows 7/8, and Windows Server 2008/2012. It is especially useful for IT administrators when real-time monitoring is needed immediately at a location. After its quick install, IT can monitor remote network information through the TruView Live Dashboard, regardless of where Virtual Pulse is installed.

Global Pulse

Global Pulse continuously monitors connectivity and performance from public sensors strategically placed around the world. With its user-friendly dashboard, system administrators can see traffic to and from their locations through any of the international pulse points to help triangulate connectivity and performance issues. Global Pulse currently monitors public network traffic from the following geographically dispersed areas around the world:

  • California, US
  • Dublin, Ireland
  • Frankfort, Germany
  • Oregon, US
  • São Paulo, Brazil
  • Singapore, Malaysia
  • Sydney, Australia
  • Tokyo, Japan
  • Virginia, US

More public pulse points will be added in the future, further expanding Global Pulse’s breadth of worldwide, real-time monitoring.

Regardless of the TruView Live location—from end-users to cloud-based servers—the TruView Live Dashboard provides in-depth reporting and monitoring of the following network conditions using its active pulses:

  • DNS Delay
  • Network Delay
  • Client Delay
  • Server Delay
  • Application Delay

With these tools, IT administrators now have real-time, end-to-end visibility. SaaS, cloud, and WAN connections are back on IT’s radar screen and deliver the accountability and proactive performance abilities they currently lack.

TruView Live is Pay-as-you-go

TruView Live, like many of the systems it monitors, is a pay-as-you-go service from NETSCOUT. That means businesses are never over- or under-provisioned when reigning in their evolving borderless enterprise. Unlike many other SaaS solutions, TruView Live is more comprehensive and affordable, with no long-term commitments. Live Adding, removing, or changing TruView Live monitoring is easily administered.

TruView Live Dashboard

Part V: NETSCOUT TruView Live Delivers

TruView Live is developed and supported by one of the oldest and most trusted names in network administration. Most network administrators are already familiar with other time-tested NETSCOUT products. Now IT administrators and datacenters can extend their reach into the borderless enterprise, meet or exceed their businesses’ performance expectations, and gain the following with TruView Live:

  • Accountability for SaaS, WAN, and cloud provider performance and SLAs
  • Network visibility from end-to-end, regardless of the application location
  • Greater end-user and employee satisfaction and productivity
  • More responsive IT and help desk with proactive monitoring and quicker problem resolution

As IT professionals and datacenters see even further growth of the borderless enterprise, they now have the power and tools to take back the control of their networks, regardless of where they reside.

References

  1. Columbus, Louis. “Roundup of Cloud Computing Forecasts and Market Estimates, 2015.” Forbes Magazine. 2015.
  2. Ibid. Forbes Magazine.
  3. Sterling, Greg. “Report: 60 Percent Of Internet Access Is Mostly Mobile.” Marketing Land. 2014.
  4. Meilach, David. “1.3 Billion Workers to Go Mobile by 2015.”
  5. Harvard Business Review. “How the Cloud Looks from the Top: Achieving Competitive Advantage In the Age of Cloud Computing.”
  6. Computer Economics. “Larger Companies Lead SaaS Adoption. 2013.

Trademarks, service marks or similar rights of products mentioned are the sole property of their respective owners.