Virtualization Technology News and Information
cPacket Networks 2018 Predictions: Enterprise Network Growth in 2018

VMblog Predictions 2018

Industry executives and experts share their predictions for 2018.  Read them in this 10th annual series exclusive.

Contributed by Brendan O'Flaherty, CEO of cPacket Networks

Enterprise Network Growth in 2018

Challenges to managing enterprise networks are different today than they were just five years ago.  Market reports from IDC and Crehan Research state that 100Gbps (100G) Ethernet is on track to becoming the dominant datacenter interconnect in the next few years, surpassing 40G by year's end. The cost benefits of 100G are leading to expanding the deployment of 100G from specialized applications to mainstream switching and routing.  That's good news.  But good news of any kind may come with precautions as well.  And that's the case with fast networks, especially considering the reality that 100Gbps is a formidable leap from the 10Gbps, and even 40Gbps, networks that transport much of the world's network traffic today. 

Let's consider some of the challenges: 

#1.  As network speeds increase, those charged with network management will face an inconvenient truth:  that problems inflicting the network multiply at higher speeds.  If you can analyze only a percentage of the traffic on a 10Gbps network, you'll be able to analyze an even smaller percentage of the traffic on a 100Gbps network.  At that speed, the source of the problem will have "more places to hide," making sampled approaches more susceptible to false positives and false negatives.

  • When you start a packet capture and analysis after the problem occurs, in most cases the packets captured may not be associated with the problem and therefore a root cause analysis is nearly impossible. The right solution is a continuous network monitoring system which will keep a track of all network activities with no impact on storage and the network.
  • Cisco Systems, Inc. predicts that global data center Internet protocol (IP) traffic will grow by 31 percent annually within the next five year
  • 100G port shipments are expected to increase dramatically (in 2013, port shipments (per units) for 10G networks and below was approximately 200,000. By 2022, 100G and above will comprise approximately 1,000,000 port shipments (per units) while 10G networks and below will remain if not decrease from 200,000. (Source Dell'Oro: JP Morgan )

#2.  As companies see quantum leap increases in the data transported across their networks, one theory says that any larger dataset will provide more actionable insights than a smaller dataset.  But practitioners are beginning to acknowledge the limitations of big data: first, that more data doesn't equate to extracting meaningful data, and second, that it may become too computationally expensive to find the data that can yield insights.

  • Big data generates value from storage and processing of very large quantities of digital information, but the challenge remains that this information cannot be accurately analyzed with traditional computing techniques/architectures
  • The McKinsey Global Institute estimates that data volume is growing ~40% per year, and will grow 44× between 2009 and 2020
  • A survey conducted by Capgemini Consulting in November 2015, only 35% of 226 IT professionals said they are far from being able to use their big data effectively. Why? Because they lacked the proper networking tools to efficiently capture, validate, and accurately analyze the data.

#3.  When a large corporate network goes down, the first notion is to suspect a security breach.  But the source could just as likely be a difficult-to-detect spike or microburst or a misconfiguration.  The problem is that the jobs of network performance monitoring and security monitoring are separate and siloed.  But, in an outage, when you silo these teams, the end result can be finger-pointing, and network problems with massive consequences go undetected or take longer to isolate.  Companies will begin to acknowledge that networking and security personnel work better as a team, just as development and operations personnel did in creating the DevOps movement.

  • Malicious actors are becoming savvier at detecting "security gaps" between teams in organizations. The challenge remains that SecOps and NetOps are often tasks with different objectives and often don't communicate as a result of a siloed corporate structure.
  • Benefits of both teams working together: a tightly coordinated security team can yield budget savings and time. For example, investing in tools and sharing security data between these teams will eliminate the need to purchase additional (and sometimes unnecessary) preventative mechanisms.


About the Author

brendan oflaherty 

Brendan serves as Chief Executive Officer of cPacket and has over 20 years of executive and leadership experience. Prior to joining cPacket, Brendan was President and Chief Operating Officer at Massana Semiconductor, where he led a successful acquisition by Agere Systems. Brendan was also Chief Operating Officer and General Counsel for Aureal Systems, a public audio technology company. Before his journey into corporate executive management, Brendan led a successful career in Silicon Valley as an attorney with the predecessor law firm to DLA Piper Rudnick, specializing in corporate law, intellectual property, and mergers and acquisitions. Mr. O'Flaherty earned his Bachelor's degree from Santa Clara University and a Doctorate in Jurisprudence, Magna cumlaude, from Santa Clara University Law School.  

Published Friday, December 22, 2017 7:45 AM by David Marshall
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
<December 2017>