Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages wha2tsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
I'm unaware of any protocol that supports L2 gateways with that capability.
Providing it would be a rather difficult to impossible, as "normal" L2 traffic doesn't see L2 hops. Consider if your L2 switches were unmanaged switches.
What you might do, if using the "closest" gateway is paramount, is use different GW IPs, and insure the hosts closest to particular GW used it. For redundancy, something like mHSRP would allow another GW to be a fallback. I.e., for example, GW A might be the active for .1, while being the backup for .2 and GW B might be the active for .2 while being the backup for .1.
Of course, what I just described would be do the trick for traffic from a host to the GW, but return traffic, i.e. traffic from GW to host, wouldn't know which GW to use that was closest to a particular host. For that, you would need different subnets. Again, you could have hosts closest to one GW use one subnet and make that GW the preferred GW for both ingress and egress while using another GW as a backup.
