Yes. I HATE the way HW vendors are using the HCL to force refreshes on customers - by refusing to certify anything greater than 18 months old. Sucks for homelabs too - trial and error and forum digging is required moreso than it should be.
18 posts • joined 17 Sep 2009
I agree that people should have more sense - but 7.0 has been out nearly 7 months and U1 is generally the place where enterprises start upgrading. Why has Cisco not got it qualified and addressed these issues yet?
I'm sure VMware will fix, but pre-April release was the time for Cisco to test this and raise the bugzilla ticket to have it fixed.
Even the stragglers of the storage and security vendors finally released 7.0 support in the last couple of months (probably initially timed to coincide with VMworld)
Data centres are warm and designed to move air very efficiently. Are they safe to visit during the pandemic?
Kubernetes is 'still hard' so VMware has gone all-in on container-related tech with expanded Tanzu, vSphere 7
Multi-cloud is their big bet
Yes it's not cheap by any means, but the extra abstraction layer gets you the ability to run cloud native workloads on premises, plus have mobility of those workloads between on-prem and multiple clouds - this is worth the money for some folks, though not for everyone.
Credit where due, VMware ease of install and lifecycle management has improved a ton the past few years, for many environments the cost is worth it for the management/ops tools. VCF is a heck of a big footprint for small shops though, they need to come up with some new Essentials like cheaper option that is limited enough to not cannibalize Enterprise sales but still has all the shiny new bits. Wavefront (now Tanzu Observer or something) and CloudHealth have been way under marketed too. Watching with interest.
How four rotten packets broke CenturyLink's network for 37 hours, knackering 911 calls, VoIP, broadband
Another 7+ owner here, I always make it two years between replacements and when I'm lucky three (don't drop it too much). Brought the wife a 7+ 256GB refurb for a good price a couple of weeks ago, would expect that to now last her two years too.
The changes are either incremental or unwanted (FaceID) this go round and while I see a surprising number of X's on planes and in peoples hands (work in Silicon Valley) unless there's a massive leap in the rear facing camera I can wait.
I was an Android person years ago, but twice burnt purchasing top of the line devices which the manufacturers decided to stop updating 18 months post my purchase. Say what you like about Apple but at least if I can make a phone last three years I know it will still run the latest software with patches.
Re: Zip code for non-US cards
The PAYG thing is insane, go to the trouble of getting your phone unlocked, manage to find a nano-SIM on a UK PAYG net, then find the only way to get more credit on it is buying a retail voucher - marked up massively compared with online. Seemed to be 4:1 last time I tried this, 2014 I think, I spent 100 quid on what online should have been 30.
Re: Sounds fair
That's precisely why when congress last discuss this back in the 90's, they delayed it pending the setup of the 'SST' streamlined sales tax arrangement, where a bunch of states agreed to pay for 3rd parties to collate and calculate and file taxes for small business. These days using a SaaS service to calculate the tax is easy and free (the states pay rather than the business).
Don't forget people are supposed to have been paying their out of state sales taxes all along, just by adding them to their tax returns at the end of the year, just nobody did.
Re: Obvious strategy is obviously missing
I think part of the problem is the obsession with VM mobility rather than application workload mobility. It's much easier to distribute load with an F5 GTM or equivalent between environments fairly seamlessly than it is to vMotion to another environment over limited bandwidth, then have all the challenges of overlay networking in order to retain IP addressing.
By all means develop Cloud Connector to move VMs around and translate between different hypervisor VM container formats, but the higher in the stack you do some of this stuff the easier it is, both to accomplish and especially to support and troubleshoot.
Disclaimer - I work for F5, I know there are other solutions available.
FCoE was trapped at top of rack until the multi-hop additions were added - it's still very limited compared to FC, so I'd fully expect FC to reign for a while longer in the core of large SANs. Not forever though, maintaining a whole separate SAN infrastructure just doesn't add up with 10 and soon 40 gig Ethernet cheaper and with QoS/lossless extensions, Brocade better stockpile that hay...
Cisco comparison is a bit silly
Yes the currently shipping Nexus blades have 32 ports of which only 8 are available in 'dedicated' non-oversubscribed mode, but that's a first generation card. There's a 32 port non-oversubscribed one due in a couple of months - still only 256 ports I know, but I doubt Arista has shipped any of these yet either. So you'd only need a pair of Nexus 7010's or one Nexus 7018 to reach 384 ports. The Cisco also has a modular backplane, and a roadmap for a lot more performance.
Granted I'm sure the Arista will still be much less expensive.