I have a question regarding the bit-level caching which Cisco WAAS performs. In our scenario, we're going to have a central site with a WAE-7326 or WAE-7341, and many (dozens) of remote sites with WAE-512 devices. Data transfer between all the sites will be very similar.
The way I understand it, is that the caching happens between pairs of devices, and the result is that chunks of data may be repeated on the central site unit for each remote site which transferred the same chunk of data.
Similar products can share the same cached data with multiple remote sites, improving scalability at the central site.
Is there any way to mitigate this behaviour? I suspect it is going to greatly reduce our cache hit rate and affect performance.
Cisco's WAAS product does indeed cache data on a per peer basis, which results in multiples of the same data being stored at the central site. This may not be a big deal for a few sites, but it is not very scalable as the hit rate for warm data is greatly impacted.
I would suggest looking at another WAN optimization solution and Riverbed's Steelhead appliance (www.riverbed.com) seems to be the the best solution. They support a universal datastore where data is only stored once and shared across all sites instead of being stored individually on a separate per peer basis. They also have over 2500 customers and an announced software client that accelerates mobile workers that connect from home or when they are traveling with their laptop.
I've finished a try and buy, and am in the process of deploying 512's to 8 edge sites, and clustering 512's in the core. I had the same concern regarding the shared cache that some of the competitors had, versus seperate data stores per pair.
According to TAC, this function was chosen for performance. Even though you will be storing duplicate data, the performance to pull this data will be quicker as you will not have one giant DB.
I'm not sure if this is a cop out on Cisco's part, but I see the point.
From the time that I spent looking at 4 other competitors, they all seemed to be marginally different as far as performance. To Cisco's benefit, it is a transparent solution aside from the CIFS piece. My understanding is that the product which used to be WAFS will soon be transparent as well. Of the products that I was considering, only one other was transparent.
There is a limit of 64 Gig of CFS data per disk on the WAE-7326 ACNS system. If you would like to use more CFS caching on the box, you can replace the large disk with several smaller disks.
This design was meant to optimize the performance of the cache. Web content is generally made up of many small objects. If you were to cache several hundred gigs of data on one drive, it would create a very large database of information and it would lose efficiency searching for the data on the disk. It is also a memory limitation.
This is documented in the software upgrade and maintenance document.
Note: Because of memory restrictions in the WAE, the maximum disk storage allocation for the cfs is 64 GB per disk. This limitation affects all WAE devices. To configure more than 64 GB of disk storage for the cfs, you must have more than one disk drive in your device.
This document will provide screenshots to outline the steps to setup
TACACS+ configuration to ACI and also the configuration required on
Cisco ACS server. Please find the official Cisco guide for configuring
TACACS+ Authentication to ACI:
Is it supported or NOT supported? It's a frequently asked question.
Before APIC, release 2.3(1f), transit routing was not supported within a
single L3Out profile. In APIC, release 2.3(1f) and later, you can
configure transit routing with a single L3Out pr...
Cisco Documents are usually accurate, but when it came to the document
on Cisco APIC Signature-Based Transactions it was slightly off the mark.
This document is for those novices to API like me who cant seem to
figure out how to go about performing signat...