Arista vs Nexus, which would you rollout?

[deanwebb]

I'm still waiting for maintenance to be activated on some gear we bought almost 3 years ago so that I can use it before the 3 year maintenance contract is up.

We had to do a complete re-architecture of one project because of the haggling between our purchasing guys and the vendors. They liked how virtual stuff, being software, came under the OPEX budget instead of CAPEX and that we could get deeper discounts on software than hardware. Out went the physical, distributed solution and in came the virtual, centralized solution.

Just a few weeks ago, we had an architecture review and the manager worked up a recommendation to start getting some physical devices to push out into our sites that don't have a big virtual presence. I'll be happy to go back to the physical, distributed solution, but I'm not going to hold my breath.

[mlan]

- Apps team with large project budget purchases network with consultants to deploy EVERYTHING. Its next week. "Hey That1guy you got a couple min to talk through how this will connect?"

Oh yeah, these "surprises" are always fun.  We have had vendors just show up in IT/data center with network appliances or equipment asking where to rack and stack.  I have also seen departments or project teams completely bid out entire network solutions including all switching without ever even thinking of talking to IT ops.  "Just make it work - contract is already signed".  This is usually due to a lack of org governance or leadership/turf wars.

[NetworkGroover]

@ AspiringNetworker

Oh the number of times I have faced any of the following or similar.

- CEO sat next to another CEO in 1st one trip. Determined we will now deploy $new_hotness from $vendor. "Signing contract today"
- Apps team with large project budget purchases network with consultants to deploy EVERYTHING. Its next week. "Hey That1guy you got a couple min to talk through how this will connect?"
- Submitted network design/BOM/PO rejected. Leadership and procurement rejected due to "$crappySwitchCo undercut bit" and this is what was purchased. "Can you have it deployed in 2 weeks?"

Its sad but I see more times project design goes as far as what switches to purchase before cutting a PO. Rest can be figured out after we get them in.

So honest question - do you ever push back with justification?  Again - I don't think I'd last in that world.  Better I stay on this side of the fence. xD

[deanwebb]

Push back with justification? If I'm lucky, this will be the higher-up's reaction:



Meanwhile, my fellow engineers are all like



DON'T DO IT MAN, THINK OF THE CHILDREN!!!

***

Back to the physical that went almost 100% virtual... the position was clear. If I wanted vendor $$$ over vendor Cheapo, then the ONLY way we'd get vendor $$$ in would be if we got their deep discounts on virtual gear, full stop, end of story. If I dug my heels in on the physical solution, it would be goodbye vendor $$$ and hello vendor Cheapo and a serious career decision for me, because I did NOT want to work with vendor Cheapo's stuff at all. I had tested it and it was a failure. We needed the $$$ stuff, but no way would our accountants pay for the hardware when the virtual stuff was so much less.

That's about the time I put "Accounting is architecture, remember that!" in my sig, I believe. Because we re-did everything in the middle of those negotiations. We got vendor $$$, but only just.

There's still talk about ripping out vendor $$$ and going with Cheapo... I did push back on insisting on $$$, but got pushed right back from above to make a big change to get that vendor over Cheapo.

I believe part of it has to be a stand we have to take so that we don't get a rep as a firm that just cuts a check for expensive stuff without a second thought. We've got a Wal-Mart type of attitude, where we know we'll drive down the purchase cost, but the vendor will make up for it on volume.

But accounting is architecture, remember that!

[that1guy15]

@ AspiringNetworker

Oh the number of times I have faced any of the following or similar.

- CEO sat next to another CEO in 1st one trip. Determined we will now deploy $new_hotness from $vendor. "Signing contract today"
- Apps team with large project budget purchases network with consultants to deploy EVERYTHING. Its next week. "Hey That1guy you got a couple min to talk through how this will connect?"
- Submitted network design/BOM/PO rejected. Leadership and procurement rejected due to "$crappySwitchCo undercut bit" and this is what was purchased. "Can you have it deployed in 2 weeks?"

Its sad but I see more times project design goes as far as what switches to purchase before cutting a PO. Rest can be figured out after we get them in.

So honest question - do you ever push back with justification?  Again - I don't think I'd last in that world.  Better I stay on this side of the fence. xD

Yes, Ive always been sure to let those directly above me know when crap like this is not OK. Now whether they listen or not is something else.

With crap like this happening these jobs were quite easy to quit and sure as hell stuff like this gets brought up in exit interviews. Stuff like this is how you run off your talent. But then again Im sure they dont care.

Sometimes though you just have to play the game and focus on what you can control. You getting to do exactly what you want in your job is not always the only deciding factor in staying or leaving.  There are many others that could come into play for anybody.

[wintermute000]

yeah just don't play the game at all. #consultantlyfe

I'm external as well so if the client insists on doing the wrong thing, go right ahead, we're gonna caveat our position then bill you anyway, and at the end of it we walk away whilst you're the ones who have to live with the consequences. Doesn't tend to happen as we position ourselves to come in right at the start so we can usually stop anything too outlandish via simply pointing to the requirements (that we drag the customer into going over for this very reason). But that's a luxury sometimes, esp. when people want to do it proposal/RFP style instead of consulting style, and esp. when mega-corps and mega-vendors are involved with their implicit nudge nudge wink wink pre-determined outcomes.

[burnyd]

@ AspiringNetworker

Oh the number of times I have faced any of the following or similar.

- CEO sat next to another CEO in 1st one trip. Determined we will now deploy $new_hotness from $vendor. "Signing contract today"
- Apps team with large project budget purchases network with consultants to deploy EVERYTHING. Its next week. "Hey That1guy you got a couple min to talk through how this will connect?"
- Submitted network design/BOM/PO rejected. Leadership and procurement rejected due to "$crappySwitchCo undercut bit" and this is what was purchased. "Can you have it deployed in 2 weeks?"

Its sad but I see more times project design goes as far as what switches to purchase before cutting a PO. Rest can be figured out after we get them in.

[NetworkGroover]

yeah just don't play the game at all. #consultantlyfe

I'm external as well so if the client insists on doing the wrong thing, go right ahead, we're gonna caveat our position then bill you anyway, and at the end of it we walk away whilst you're the ones who have to live with the consequences. Doesn't tend to happen as we position ourselves to come in right at the start so we can usually stop anything too outlandish via simply pointing to the requirements (that we drag the customer into going over for this very reason). But that's a luxury sometimes, esp. when people want to do it proposal/RFP style instead of consulting style, and esp. when mega-corps and mega-vendors are involved with their implicit nudge nudge wink wink pre-determined outcomes.

Ugh - yes!  That comment about proposal/RFP instead of consulting... man!  "Give me what your solution is lined up against this Vendor XYZ BoM and no other info - I wanna see pricing."  - this drives me INSANE.

[deanwebb]

That proposal/RFP method is primarily asking for the other party to act as a staffing agency, not as a value-adding consultant. Basically, it's, "We need 15 level 1 guys, 9 level 2 guys, 3 level 3, one level 4, and two service managers. Level 1-2 guys should all be CCNAs familiar with Riverbed and the 3/4 guys need to be certified by the vendor. Also CCIEs. How much will that cost us and how soon will you have them delivered?"

I hate, really hate, stuff like that. Either hire real consultants to do the job or real people to work for your firm. Don't go through some staffing solution crap to fill personnel reqs while keeping payroll low.

[NetworkGroover]

Hmmm - think I misunderstood what was said here.

[NetworkGroover]

@ AspiringNetworker

Oh the number of times I have faced any of the following or similar.

- CEO sat next to another CEO in 1st one trip. Determined we will now deploy $new_hotness from $vendor. "Signing contract today"
- Apps team with large project budget purchases network with consultants to deploy EVERYTHING. Its next week. "Hey That1guy you got a couple min to talk through how this will connect?"
- Submitted network design/BOM/PO rejected. Leadership and procurement rejected due to "$crappySwitchCo undercut bit" and this is what was purchased. "Can you have it deployed in 2 weeks?"

Its sad but I see more times project design goes as far as what switches to purchase before cutting a PO. Rest can be figured out after we get them in.

So honest question - do you ever push back with justification?  Again - I don't think I'd last in that world.  Better I stay on this side of the fence. xD

Yes, Ive always been sure to let those directly above me know when crap like this is not OK. Now whether they listen or not is something else.

With crap like this happening these jobs were quite easy to quit and sure as hell stuff like this gets brought up in exit interviews. Stuff like this is how you run off your talent. But then again Im sure they dont care.

Sometimes though you just have to play the game and focus on what you can control. You getting to do exactly what you want in your job is not always the only deciding factor in staying or leaving.  There are many others that could come into play for anybody.

Fair.

[NetworkGroover]

Push back with justification? If I'm lucky, this will be the higher-up's reaction:



Meanwhile, my fellow engineers are all like



DON'T DO IT MAN, THINK OF THE CHILDREN!!!

***

Back to the physical that went almost 100% virtual... the position was clear. If I wanted vendor $$$ over vendor Cheapo, then the ONLY way we'd get vendor $$$ in would be if we got their deep discounts on virtual gear, full stop, end of story. If I dug my heels in on the physical solution, it would be goodbye vendor $$$ and hello vendor Cheapo and a serious career decision for me, because I did NOT want to work with vendor Cheapo's stuff at all. I had tested it and it was a failure. We needed the $$$ stuff, but no way would our accountants pay for the hardware when the virtual stuff was so much less.

That's about the time I put "Accounting is architecture, remember that!" in my sig, I believe. Because we re-did everything in the middle of those negotiations. We got vendor $$$, but only just.

There's still talk about ripping out vendor $$$ and going with Cheapo... I did push back on insisting on $$$, but got pushed right back from above to make a big change to get that vendor over Cheapo.

I believe part of it has to be a stand we have to take so that we don't get a rep as a firm that just cuts a check for expensive stuff without a second thought. We've got a Wal-Mart type of attitude, where we know we'll drive down the purchase cost, but the vendor will make up for it on volume.

But accounting is architecture, remember that!

Haha just revisited this thread and saw this - made me chuckle

[deanwebb]

Haha just revisited this thread and saw this - made me chuckle

Mission accomplished.