Greg Copeland
Director, Business Development

VIAVI Observer & Ixia CloudLens: 6 Steps to Regaining Control in the Cloud

February 6, 2018 by Greg Copeland

Co-authored by Steve Brown of VIAVI and Greg Copeland from Ixia.

Joint solution empowers IT with visibility into cloud assets to ensure optimal end-user experience.

Today business leaders opt to migrate services to public clouds without consulting or even notifying IT of the change, until an issue arises. While this has become common, IT teams are still responsible for finding and fixing problems, remaining accountable for performance, security, compliance, and return on investment (ROI).  IT is also still responsible for user experience, but has lost control over some critical decisions.

Nor does the challenge of maintaining control end with initial cloud deployment. At its recent Data Center Conference, Gartner asked engineers about their biggest cloud monitoring challenges. More than 50 percent said they were “blind to what happens in the cloud” while 32 percent cited visibility gaps and the majority felt the data shared by cloud providers did not meet their needs.

How can IT regain control in the dynamic world of cloud? 

The good news is that, despite the move to cloud, monitoring best practice requirements do not change all that much. Cloud providers do offer some limited metrics about hosts, along with high-level metadata about usage and health, though this insight alone is not enough. Deeper analysis and forensics, application performance monitoring (APM), and security solutions all require access to packet-level data, and this can still be achieved with the VIAVI-Ixia joint solution.

VIAVI and Ixia have teamed up to create a six-step approach to capturing and analyzing the same quality of data IT utilized in the physical world. On the surface, the steps mirror strategies to pinpoint problems, devise fixes, and improve operations now.  In reality, new capabilities are needed throughout the process:

STEP 1 - Access to data must be extended to extract packet-level data from the cloud 
STEP 2 - Intelligent filtering of data to APM solutions must occur dynamically as instances are created 
STEP 3 - Data storage strategies must minimize excessive cloud usage fees 
STEP 4 - Troubleshooting needs to begin within the cloud, versus having to backhaul all data 
STEP 5 - Detailed analysis involves new, cloud-specific metrics 
STEP 6 - Measuring success hinges on before and after evaluation of migrated services

Together, VIAVI and Ixia equip IT to follow these steps to capture data and make informed decisions about working with cloud providers, meeting service-level agreements (SLAs), and improving the user experience. 

Steps 1 and 2: Data Capture

In 2017, Ixia launched its cloud-native visibility solution, CloudLens Public, to enable first-of-its-kind access to packet-level data residing in the cloud.  A cloud-native solution that scales automatically and dynamically as cloud instances are created or destroyed, CloudLens Public introduces a software as a service (SaaS) approach to visibility that saves users time and money. 

CloudLens sensors are installed within cloud workloads to access rich metadata about where services are running, associated Internet Protocol (IP) addresses, and memory and central processing unit (CPU) utilization. A drag-and-drop interface is used to define packets to be forwarded between two types of groups:

  • Production groups such as email, voice, and domain name system (DNS)
  • Tool groups such as APM, network performance management (NPM), and intrusion prevention systems (IPS)

The CloudLens interface also allows for filtering paths between the two. New instances are automatically added to the right source groups based on characteristics and traffic is filtered and load balanced across the prescribed tool groups.

The Ixia sensor-based approach allows CloudLens to bridge visibility gaps across multi-cloud environments by working with Amazon Web Service (AWS), Azure, Google Cloud, and other popular provider solutions with no additional effort required to define new security policies. Intelligent filtering eliminates the need to send data to tools that do not need to see it, which optimizes the use of analysis tools and consumption-based cloud services.

Step 3: Storage

Once you capture this data, what do you do with it all? Providers charge for processing, storage, and import or export on a usage basis, so best practice would be to keep collective data in the cloud if possible. The joint Ixia–VIAVI solution allows metadata to be used for real-time monitoring in the cloud and packet-level data to be downloaded as needed to reconstruct scenarios and perform forensics.

Steps 4 and 5: Analysis and Troubleshooting

The data shared by cloud providers is, unsurprisingly, cloud-centric, focusing on compute, network, and storage performance metrics of cloud resources. However, the job of IT does not stop there.

To manage deployments, IT needs a clear picture of the user experience with access to similar metrics and network traffic that engineers would have on-premises. Metadata from providers can be used to start triaging symptoms, but root-cause analysis and security forensics require the actual packets. 

Once data from CloudLens reaches the VIAVI Observer Apex solution, users can spot trends, such as best and worst-performing applications, then drill down to isolate and troubleshoot issues. For example, if a web-based application performs poorly, analysis might show an abundance of “page not found” errors or corrupted pages and reveal the cause to be issues with a particular server. 

VIAVI Observer Apex features guided out-of-the-box workflows designed to promote rapid isolation and resolution of issues within three clicks. Does the problem lie within the on-premises infrastructure or the cloud? Is it regional or widespread? It is unique to one provider? Are the delays arising on your side or the providers’?

The VIAVI multi-segment analysis helps pinpoint where along the path conversations are failing in multi-provider environments. By understanding where breakdowns are occurring and the dependencies between applications, compute, and storage resources, IT can manage provider relationships more effectively.

Step 6: Measuring Success

At the end of the day, IT helps business leaders understand what constitutes “normal” pre- and post-migration, as well as the success of cloud migration over time. The latter may be reviewed in terms of reliability, agility, and efficiency.

What percent of SLAs are we meeting? Is the Help Desk getting more calls on migrated services or fewer? Has any element of the service been left behind?

The metrics for measuring cloud infrastructures include those used to evaluate systems on premises, including response time, round-trip time, data transfer speed, bandwidth utilization, resource consumption, and corrupted or invalid transaction requests, along with new requirements for breaking down paths from data centers and key offices to cloud providers. 

Apex also helps in flagging anomalies, such as spikes in demand, so IT can alert business leaders to potential cost overruns and determine ongoing requirements for monitoring and analysis.

Why the Joint Solution?

Together, Ixia and VIAVI provide:

•    Access to critical data wherever it resides
•    Efficient, intelligent distribution of data to the right tools
•    Fast, intelligent analysis of data
•    Near instant insight into network and application performance on premises or in the cloud 

Companies still need to educate business groups about engaging IT during initial planning stages, but IT is once again empowered to safeguard the end-user experience and corporate reputation. 

Watch the webinar now on demand or download the new joint solution brief.