Around the Storage Block
StorageExperts

Make data center networks smarter with actionable insights featuring “What Just Happened” telemetry

Network downtime has always had a big impact on business and revenue. But with an increase in the number of mission-critical workloads running on denser and faster datacenter infrastructures, businesses today stand to lose more than in the past.

Intelligent fabrics for data center_Blog_shutterstock_279683456.pngThe greatest challenge in reducing downtime is identifying the root-cause of problems. Likewise, incorrect root-cause diagnosis can lead to a multi-fold increase in time-to-resolution and additional loss. With a direct correlation between network visibility and time to resolution, network monitoring and analytics tools must keep pace. However, as data center network speeds rapidly increase, legacy monitoring techniques leave room for concern.

 Challenges with legacy telemetry

The sheer volume and variety of data over a typical network is growing all the time, creating a complex and noisy environment. Increases in network speeds also mean an increase in data that must be analyzed in shorter periods of time. Traditional network monitoring approaches and tools use a central collector for telemetry data. These traditional approaches present a growing challenge to the transportation, processing, and storage of telemetry data as even a small network can generate petabytes of data per day.

The network itself contains a wealth of information that can be used for smart decisions and to determine root-cause in the presence of a problem. So why can’t the ethernet switch -- which has visibility to each data packet that crosses the network -- be used to gather telemetry? The difficulty is that, in general, commodity ethernet switches can’t provide the critical telemetry data needed for root cause problem identification or issue resolution. This forces network administrators to process massive amounts of telemetry data centrally, creating an artificial bottleneck to problem resolution.

What just happened (WJH) telemetry

Imagine having easy access to all the information you need to avoid potential network and service disruptions before they happen. HPE M-series ethernet switches have the capability of providing rich, contextual and actionable insights on a variety of data. This includes layer-1 through layer-4, ACLs and buffer occupancy. Powerful agents collect and store network events to provide visibility and insights for critical questions in problem solving, answering who, what, when, why and where queries. We call this ‘What just happened’ and it is now available on every M-series switch platform that is running any network operating system, and can be integrated and extended with third party and open-source tools.

An improvement over traditional streaming and in-band network telemetry, WJH skips over much of the time-consuming data collection and manual sifting for network problems. With WJH, traffic inspection, filtering, and issue identification are done by the switch where the network context is readily available. As a result, relevant issue-related data is captured. For example, when packets are dropped, WJH implements the appropriate drop counter, as well as capturing the relevant header data from the packet so a more thorough analysis can be performed. With all of the relevant data at your fingertips, WJH helps network administrators to dramatically reduce meantime to issue resolution. Additionally, this information can also be used to provide insights to help improve resource utilization and capacity planning.

How does it work?

What Just Happened is an advanced streaming telemetry technology that provides real-time visibility into network-based problems. What Just Happened goes beyond conventional telemetry solutions by providing actionable details on abnormal network behavior. Traditional solutions try to extrapolate the root cause of network issues by simply analyzing network counters and statistical packet sampling. What WJH eliminates the guess-work from network troubleshooting by leveraging the unique hardware capabilities built into the M-series switch ASICs to inspect packets at multi-terabyte speeds. WJH inspects packets across all ports at line-rate and at-speeds.

Figure 1: Build intelligent Data Center Fabrics with HPE M-seriesFigure 1: Build intelligent Data Center Fabrics with HPE M-series

 

For storage administrators, troubleshooting network and SAN issues has always been a challenge. The combination of WJH and HPE Network Orchestrator enhances HPE M-series solutions and simplifies SAN management and troubleshooting by providing advanced diagnostics that help keep everything running smoothly without performance degradation. With WJH, one can solve problems and easily debug severe network issues in minutes.

Summary

The HPE M-series ethernet switch family delivers the highest levels of performance at all the relevant speeds, including 1, 10, 25, 40, 50 and 100Gb/s. They have the lowest latency of any mainstream switch, featuring zero packet loss (no avoidable packet loss, for example due to traffic microbursts), and delivering that performance fairly and consistently across any packet size, mix of port speeds, or combination of ports.

Traditional network visibility methods fall short as datacenters adopt 25G, 50G and 100G ethernet speeds. What Just Happened provides an extensible and actionable telemetry that can help operators dramatically reduce mean-time to issue resolution, and improve infrastructure uptimes at even higher data rates. WJH leverages M-series Spectrum silicon level capabilities to provide granular, contextual and actionable telemetry information on datacenter networks.

What Just Happened can directly be used on the switch platforms via CLI. Alternatively, the WJH functionality is also available through a Web UI.

Headshot Faisal_400x400.jpgMeet Around the Storage Block blogger Faisal Hanif, Product Management, HPE Storage and Big Data. Faisal is part of HPE’s Storage & Big Data business group leading Product Management & Marketing for next generation products and solutions for storage connectivity, network automation & orchestration. Follow Faisal on Twitter @ffhanif 

 

 


Storage Experts
Hewlett Packard Enterprise

twitter.com/HPE_Storage
linkedin.com/showcase/hpestorage/
hpe.com/storage

For more information, please check out the following:

Blogs

White papers

Video

TekTalk webinars

Web

 

0 Kudos
About the Author

StorageExperts

Our team of Hewlett Packard Enterprise storage experts helps you to dive deep into relevant infrastructure topics.