Page 10 - Cisco Tribune Q2 2015

l l l
10 2nd Quarter 2015 NEWS PRODUCTS PEOPLE EVENTS
Telecom Reseller: Cisco Tribune

form the WAN/LAN monitoring.
virtualization technology to consolidate
virtualized automated network can happen end result will be analyses that do not offer
Network and data center monitoring data many network equipment types onto in milliseconds. If left unanalyzed, the useful actionable results.
deinitely qualiies as big data. It continues industry standard high volume servers, data value may decrease to where it is only
to increase the volume of data produced. switches and storage, which could be historical data. Then the data cannot be What do we need?
The monitored data also changes rapidly in located in datacenters, network nodes and used to deal with real time situations.
Software-deined networking is a
milliseconds and even microseconds, and in the end user premises.” If the ETSI
* Variability--Data creation does not networking technique that allows network
can change faster than humans can analyze can standardize NFV and its monitoring follow a nice smooth creation pattern. administrators to manage network
and respond.
data, then we may not have the same Large data production may be produced by resources. This is done by separating the
multivendor approach of SDN. If not, then unexpected events (trafic bursts, resource resource that makes decisions about where
Big Data as a Tool
we need standards here as well.
failure, and security attacks) as well as trafic is sent from the resources that
In Big Data: A Tool, Not an Answer, I described ive factors of big data that A third entry into the confusion is the service provider community. Nearly all periodic events.
* Many Formats--It would be nice if all forward trafic to the selected destination. We do not have one but several SDN
produce challenges for anyone who wants enterprises use some provider(s) in their the data was in a common format. Vendors approaches on the market. We need some
to process big data and generate useful and networks. The providers may be reluctant add proprietary data extensions to make agreement on the data generated by
timely results:
and therefore limit the data available for big their products more attractive by adding different vendors and standardization of
* Quantity--The amount of data produced data analysis. They may also add their own more collectable information. The variety format and data content.
by a wide range of network resources keeps formats and content information.
of formats and data extensions is already
increasing. Data comes in both structured What we need is a holistic approach to
signiicant. This makes analysis that much The Network Functions
and unstructured forms. It depends on the big data network monitoring analysis. The more complicated.
Virtualization Working Group of the
vendor of the resource.
fragmented condition of today means that * Many Sources--Attempting to connect, European
* Delivery Speed--The rate of data the enterprise will have to focus on one link, match, and transform the data is quite Telecommunications Standards Institute
delivery stimulates rapid processing of resource vendor to reduce the variety of data a task. If correlations cannot be made, data (ETSI) states that NFV “aims to address.
data. Trafic and resource changes in a
elements with varied formats and content. relationships will be fragmented and the
problems by leveraging standard IT
This will still not resolve the monitoring
data difference from the providers.
Finally, many big data tools are focused
on human behavior and consumer data.
The tool vendors will have to expand
their products to handle the real-time fast
response requirements of the network staff at the enterprise.

Conclusion
In Big Data: A Tool, Not an Answer, I wrote,
“Big Data is nothing unless the data is
properly digested, correlated, matched, and
transformed across systems. A problem
for the [networking] staff in charge of Big
Data is that most of the collecting systems
are not connected--nor are they designed
to be connected. It appears that analyzing
collected data will require training,
experience, new system connections, and a
lot of feedback to see if the right data was
collected and if the right questions were
asked.”
The end results of analyzing the
monitoring data will be useless and possibly
dangerous to the health of the network and
therefore detrimental to the enterprise.
An extensive list of various types on
monitoring tools has been created at the
Stanford Linear Accelerator Center http://
www.slac.stanford.edu/xorg/nmtf/nmtf-tools.
html#intro
Ed Note: the original piece has several
links not reproduced herein. See online
version at tinyurl.com/on6sj8z for access to
all links. n

Cisco
©
TRIBUNE

THE INDEPENDENT REPORT ON THE CISCO TELEPHONY MARKET
www.ciscotribune.com

is a production of

SUBSCRIPTIONS

Or change address, unsubscribe
Call: 800 667-8965 •
Fax 360 260-9726
Web: www.ciscotribune.com
Email: publisher@usernews.com
ADVERTISING

Call: 360 260-9708 •
Toll Free: 800 667-8965

Fax: 360 260-9726
Email: publisher@usernews.com

p1-8_2ndQtr_Cisco_Tribune_2015.indd 10
04/06/2015 10:34