Community
Command Central
MVP Program
Monthly Mission
Blogs
Groups
Events
Media Vault
Products
Observability
Network Management
Application Management
IT Security
IT Service Management
System Management
Database Management
Content Exchange
SolarWinds Platform
Server & Application Monitor
Database Performance Analyzer
Server Configuration Monitor
Network Performance Monitor
Network Configuration Manager
SQL Sentry
Web Help Desk
Free Tools & Trials
Home
Products
Network Performance Monitor (NPM)
Utilization charts and frame-relay
rkendrick
I am tasked with sizing appropriate links for our frame-relay network and I'm using Orion to monitor the serial and serial sub-interfaces on our Cisco routers. When I look at the interface min/max/avg utilization charts for 1, 7 or 30day periods I see utilization but my question is how does one determine what is appropriate. Example: A hub site has a T1 access port and remote has a 256 access port speed. The PVC between these two sites has a CIR of 64k. Looking at the hub site's PVC graph between these two, if I see an TX average utilization of 60k over 7days and a RX average of 14k over 7days, am I to assume I'm safe with the 64k CIR or do I need to add both the TX+RX together to determine if I'm exceeding my frame-provider's CIR limit. How should peaks come into play on my decision. The links are all full-duplex in most of our locations. What difference would I need to considering if there is a half-duplex interface in the mix? Thanx.
Find more posts tagged with
Accepted answers
All comments
ggardner1
I am lucky enough to be in a company that has a subscription with the Gartner Group and have found their research to be invaluable on many occasions. I happened to remember an article I had read a few years back that had helped me with this issue. In general, they recommended that you design a Frame network based on port speed within the U.S. and on CIR when outside the U.S. I recommend that you stop by their site annd take a look at their research. You can purchase individual articles if you can't get a buy-in on a subscription. If someone wants to know the title of that article just let me know.
Generally there is so much overcapacity in the U.S. that even when you go over your CIR your traffic will still be passed. I have many sites with a full T1 Port, only a 128K cir and they run 30-50% utilized on a regular basis. When I regularly see utilization over 60-70% then I generally know that its time to upgrade a link. I also use response time as a guide and have a trigger that alerts us on our cell phones whenever a remote router gets to 200ms or higher. The way I look at it, its not the lines that matter but the user experience and the response time generally can give you insight to that. Solarwinds can't do it, but I find that looking at FECNs and BECNs on the router can also give you a good indication of problems. I have 1 site that gets high FECNs and BECNs and have had to up their CIR a few times to compensate, but the other sites run fine.
International is annother story, and I design my International lines based on CIR. I generally put my International CIR at 50-100% of port because there is much more competition for bandwith.
Thanks,
Greg Gardner
BryanBecker
I would also say most US providers offer 0 cir FR connections. They all say they can run at port speed/0 cir and not drop packets.
You can also deploy a QoS (Quality of Service) strategy that can prioritize applications on those lower speed links.
BB
Quick Links
All Categories
Recent Posts
Activity
Unanswered
Groups
Help
Best Of