Re: [j-nsp] What is an acceptable amount of latency for traffic routed through an SRX cluster?

2012-01-10 Thread Mark Kamichoff
Hi Morgan - On Mon, Jan 09, 2012 at 03:23:57PM -0800, Morgan McLean wrote: > Its an SRX3600 cluster, with no traffic traversing the fabric > connection, so its all being contained on one chassis. These are just > standard ICMP packets between two linux hosts on different subnets. By ICMP packets

Re: [j-nsp] What is an acceptable amount of latency for traffic routed through an SRX cluster?

2012-01-09 Thread Jared Mauch
I feel compelled to share this link: http://www.snookles.com/slf-blog/2012/01/05/tcp-incast-what-is-it/ Just in the event you haven't seen it or looked deeper. Jared Mauch On Jan 9, 2012, at 8:13 PM, Morgan McLean wrote: > We're running over a terabyte in membase, but thats besides the point

Re: [j-nsp] What is an acceptable amount of latency for traffic routed through an SRX cluster?

2012-01-09 Thread Morgan McLean
We're running over a terabyte in membase, but thats besides the point. Question still stands :) Morgan On Mon, Jan 9, 2012 at 4:56 PM, Joel jaeggli wrote: > On 1/9/12 16:28 , Morgan McLean wrote: > > Yes, we are using it for security purposes. Why would I spend so much > money > > on a box that

Re: [j-nsp] What is an acceptable amount of latency for traffic routed through an SRX cluster?

2012-01-09 Thread Joel jaeggli
On 1/9/12 16:28 , Morgan McLean wrote: > Yes, we are using it for security purposes. Why would I spend so much money > on a box that is so limited in throughput due to its various fw inspection > overhead? > > I am running two 3600's that connect via 10GE to a couple core 8208 EX > switches, which

Re: [j-nsp] What is an acceptable amount of latency for traffic routed through an SRX cluster?

2012-01-09 Thread Morgan McLean
Yes, we are using it for security purposes. Why would I spend so much money on a box that is so limited in throughput due to its various fw inspection overhead? I am running two 3600's that connect via 10GE to a couple core 8208 EX switches, which then multihome down to top of rack switches. The 3

Re: [j-nsp] What is an acceptable amount of latency for traffic routed through an SRX cluster?

2012-01-09 Thread Phil Mayers
On 01/09/2012 11:23 PM, Morgan McLean wrote: Its an SRX3600 cluster, with no traffic traversing the fabric connection, so its all being contained on one chassis. These are just standard ICMP packets between two linux hosts on different subnets. I assume you are using these as a firewall, not ju

Re: [j-nsp] What is an acceptable amount of latency for traffic routed through an SRX cluster?

2012-01-09 Thread Morgan McLean
Its an SRX3600 cluster, with no traffic traversing the fabric connection, so its all being contained on one chassis. These are just standard ICMP packets between two linux hosts on different subnets. Thanks, Morgan On Mon, Jan 9, 2012 at 3:10 PM, Joel jaeggli wrote: > srx covers at least three

Re: [j-nsp] What is an acceptable amount of latency for traffic routed through an SRX cluster?

2012-01-09 Thread Joel jaeggli
srx covers at least three different hardware architectures... An srx 5800 in a publicly available testing can do sub 300usec forwarding on small packet workloads. what srx we're talking about would probably set the expectation a bit better. joel On 1/9/12 14:40 , Morgan McLean wrote: > In our

[j-nsp] What is an acceptable amount of latency for traffic routed through an SRX cluster?

2012-01-09 Thread Morgan McLean
In our switch domain we have typically .1 ms of latency, however for traffic being routed through a gateway on the SRX to another vlan, the latency jumps up to about .5 or .6ms. Is this normal? Is this high? My boss is putting pressure on me because he says this added latency is stressing some cach