Skip to content

Another submarine cable cut, another excuse for my latency

Networking by steveipw 2 replies 115 views
8 #1

Third one this quarter. SEA-ME-WE something or other got dredged again, and now my Plex server in the basement is "unwatchable" according to my roommate. Never mind that I'm pulling from a GreenCloudVPS VPS in their "premium" tier — https://greencloudvps.com.

For that money you get:

  • 1 vCPU "burstable" to 2.4GHz
  • 512MB RAM
  • 20GB HDD (not SSD, HDD, confirmed)
  • 1TB bandwidth on 100Mbps shared

But sure, it's definitely the cable cut adding 40ms. Not the fact that the CPU throttles to 800MHz after 30 seconds of load. I ran iostat during a "buffering" episode and the disk was basically idle. CPU pegged at 100%, load average 8 on a single core.

Anyone else using cable news to hide their bargain-basement hardware sins?

zfs send | zfs receive. repeat.
#2

Cable cuts are always the excuse I see this every week my users blame the network when its their $2 vps for that money you get what you pay

I checked my logs last time someone opened a ticket about "routing issues" their container was swapping to disk not even a network problem

Emojis dont lie

#3

Someoneelses infrastructure someoneelses problem until it becomes your problem. The Cable cut industrial complex has trained Users to blame the Ocean instead of their own purchasing Decisions[1].

I have seen this pattern at three Employers now. The "premium network" marketing from budget Providers creates a learned helplessness where any Performance issue must be External. Never the Oversubscribed node never the ContendedDisk never the FairUsePolicy you agreed to by existing.

Footnotes:
[1] see also: "the Datacenter had a power issue" when the actual cause was a single point of failure in your own Rack

Post a reply

You need an account to reply. Log in or register to join the conversation.

Post reply Preview Save draft