What Is Data Center Analytics?

Why it matters

Organizations are deploying applications in multiple public and private clouds, with more applications than ever. There are also more different classes of people and machines using these applications. 

As a result of containers, microservices, and serverless, developers are constructing these highly distributed application constructs with workload tiers and data services spread across hybrid IT spanning on-premises data centers and multiple public clouds. Because of these trends, multicloud data center operators are facing serious challenges, including:

• 43% of skilled IT staff time is spend on troubleshooting in break-fix mode.
• The undisputed No. 1 cause of network outages is human error, leading to unplanned downtime.

These issues require network operators to have a high level of domain expertise and the ability to correlate complex IT environments to prevent or fix issues while upholding the infrastructure uptime to honor Service-Level Agreements (SLAs) with minimum disruptions.

So, what's happening?

In short, IT practitioners don't have the capacity to address these issues without sufficient automation tools. Here's why:

Tool fragmentation

• Too many tools addressing siloed visibility use cases
• Some are old, and some are expensive
• Different protocols/mechanisms

Limited insights 

• Low data fidelity that is not actionable
• Lack of data correlation; don't get the full picture
• No dataplane visibility

Reactive posture

• Inconsistent API architecture
• Specialized knowledge required
• Difficult to find root-cause issue, and often too late to react to it

But the tools we currently have are inadequate for today's complex networking environment.
The tools we do have are fragmented.
Which means the data we have offer limited insights.
Which leaves networking teams in a reactive posture, which is the opposite of what we want and what the business is demanding.
So how can we move forward? How can we maximize uptime, align with the business, and still have capacity for new projects and initiatives?

Our strategy and approach

To be successful, IT needs to be in a strategic partnership with business. Without this, it's impossible to efficiently help enable the changes necessary to help enable business growth. Cisco believes analytics enable IT professionals to turn raw data into actionable insights that they can use to drive business growth.  When IT practitioners move to a proactive operations approach for their data center, both sides win.

As a key element of our Intent-Based Networking (IBN) strategy, we're delivering a powerful combination of data center analytics and automation capabilities — within and across domains — to help our customers simplify their network operations and attain the insights and assurance needed to continually evolve them.

• The first step was pulling critical telemetry information out of the data and control planes and making it available to the analytics layer. We did this through silicon innovation, turning every networking device into a sensor.
• Stitch together network, security, and application analytics to provide a Single source of truth for IT operations teams and a unified view across data center, campus, WAN, branch office, and cloud environments.
• Provide AI/ML-based decision support tools for a range of common operations activities such as upgrade planning and software release guidance, proactive service level monitoring, smart troubleshooting based on graph-based search.
• Extend to cloud-based analytics offers and mobile phone dashboard option.

Benefits to business:

1. Highest Operational Uptime and Outage Mitigation to meet SLAs/SLOs

2. OpEx (Operational Expenditure) Optimization and IT Strategic Agility Enhancement

3. Security Compliance and Assurance

Benefits to IT:

1. Faster remediation of issues while increasing agility

2. Allow engineers to focus on mission critical work

3. Greater confidence and less risk in operating your network

According to latest Gartner report "2019 Strategic Roadmap for IT Operations Monitoring."

To do this, IT needs a way to transform and get past install, provisioning and segmenting. In other words, IT needs a way to make "Day-2 Operations" easier. They need to be able to: 

• analyze every component of a data center to first
• assure business intent;
• guarantee reliability;
• and finally, identify performance issues in a network before they happen.

IT practitioners can do this through intent-based networking (IBN) and automation, which allows them to:

• predict the power of their entire data center;
• gain visibility for hybrid and multiclouds as well as SD-WAN;
• analyze traffic and apps across hybrid clouds;
• identify traffic patterns;
• plan security and troubleshoot effectively;
• optimize application performance by eliminating bottlenecks,
• and more.