Sometimes you just want something now! Look, I appreciate that some things take time but in the fast paced world of technology sometimes I need it now! Like yesterday!
In the world of network visibility there are new solutions in development that promise greater insight, greater visibility, predictive analytics and so on but is visibility really the issue? Or is our ability to consume the data we already have more the issue? The problem is really twofold so what is being done to solve this issue?
From experience, gaining network visibility especially for replay of events usually involves running several time consuming reports. The way in which we troubleshoot irregular operations today often leads to much delay in resolution. Using what we’ll just refer to as an ‘exotic seven figure tool’ I have personally waited as long as 30 minutes to build connection table for replay of events.
I was recently highly impressed with a presentation by one of the vendors at Network Field Day 12, Kentik. In summary, what I perceive that Kentik has built is a highly optimized big data back end which is capable of consuming, organizing and presenting the data in near real-time fashion. Kentik continually injests data from data sources such as NetFlow, sFlow, IPFIX, SNMP and BGP and can return a wealth of data based on this information in near real time for the last 90 days or longer!
Check out the following data analysis video for more detail.
In short, I was highly impressed and I believe that tools like this are going to be critical as the complexity and volume of data that operators need to sort through is only going to continue to increase.
If you are like me trying to deal with managing complex infrastructures with little to no downtime I need more data, more often, more readily accessible. I need it all and I need it now! If so Kentik is definately worth taking a look at.
Disclaimer: I don’t get paid to write any of this stuff, I just get to hang out with a lot of really smart people for a week which is worth a lot so … assume what you want.
I have some limited background with Solar Winds. I used them in the past in much smaller networks for network management and their solution is certainly well suited for that. One thing I really like about SolarWinds is that they know exactly who they are and what they do. They are not going to be a best of breed solution in any area but they are certainly a viable solution in just about every area of network management.
During their presentation they featured two newer capabilities of their platform, logging & event correlation (aka SIEM – Security Information and Event Management) and wireless heat maps (new in v11.5 Beta). My quick assessment of this was it was both of these features were basically Splunk and Cisco Prime light.
Solarwinds SIEM information: http://www.solarwinds.com/log-event-manager.aspx
NPM v11.5 Beta info: https://thwack.solarwinds.com/message/239508#239508
I was especially impressed with the logging and event correlation piece as this is not an easy area to be competitive in but I think Solar Winds has nailed exactly what they needed to do. The provide a very solid offering in this area which will be a huge benefit to smaller organizations who look to Solar Winds for end-to-end network management.
One key point on the wireless management piece is that they currently only support Cisco APs. Hopefully support for other wireless products will be rolled in soon but I am guessing the vast majority of Solar Wind’s target market are going to be Cisco shops.
I was pleased with SolarWinds presentation. They know who they are and they have fun with it. They are doing exactly what they need to do to continue to dominate in the market they service, adding features that were probably out of the reach for smaller organizations prior.
I really didn’t know what to expect from NetBeez. As I said in my pre-NDF impressions I thought the idea of having to deploy an additional device to perform network analytics was redundant. I was thinking there had to be another way.
After seeing NetBeez presentation I had a change of opinion. As I finish this draft I am on a flight home but I plan to setup my Beez (a NetBeez agent that’s what they call a box) as soon as I get home and get some actual time with it. I will have more impressions to write about after that obviously.
The first impression I had was that this was a solution still in its infancy but the presenters were very eager to hear our feedback. I provided some feedback suggesting that a PoE option would be immensely useful. I was elaborating on that thought when I dunked my new Macbook in coffee which brought a sudden end to that.
To summarize the NetBeez device is really just a Rasberry Pi device that reports back to a cloud service. The devices can be used to monitor a number of user experience metrics and can also be used to build iperf sessions between individual NetBeez devices to assist with troubleshooting site-to-site connectivity issues. The management console appeared to be very well built and was definately aesthetically pleasing.
The point I was making though was that I could see far more potential that deploying a single NetBeez device to each site to effectively just monitor WAN performance in what amounts to elaborate IP SLA. I can see deploying a device to each IDF to monitor both LAN and WAN performance simultaneously.
Further, they announced to us a new wireless NetBeez model which I thought had immense potential, especially when coupled with PoE. I could definitely see scattering a few NetBeez devices around office complexes and especially outdoor areas to monitor WiFi signal strength and performance from the user’s perspective.
All in all, the presentation was good but the solution is very early in development so I am sure they have more features to be developed. Bob McCouch did a much more complete write up on NetBeez and if what I have stated here piqued your interest then I highly suggest you check out his blog: