Just curious, as this applies to the larger pools - do you have failover strategies in place to deal with DDoS attacks? As GigaHash capacity pools consolidate hashing power, you inevitably become a bigger target.
I don't have anything against your enterprise, just seems like a concern you'll have to deal with eventually.
Hi!
Thanks for your intrest!
This is a concern we took into account from the beginning and indeed we have several strategies in place to deal with various types of ddos but im going to be a bit cagey answering your question for a couple reasons.
1) Im originally a network engineer by trade and have dealt with large scale ddos attacks before.
2) By disclosing too much about our strategies i possibly make our strategies less effective.
3) By saying something along the lines of "we are ddos proof" or "we cannot be taken down" you simply reveal your ignorance about how certain types of ddos work and why and how they cause a denial of service. You also possibly make yourself a target where perhaps previously you would not have been before you opened your mouth and started saying silly things.
I wont use this as an opportunity to school anyone about distributed denial of service attacks as there is much material to be read about this on the net already and i cant hope to cover it in a short post but its important to mention that certain types of ddos cannot be protected against at all (bandwidth based attacks with the sum of the attackers bandwidth exceeding that of your own backbone uplinks). Given enough attacking client machines ("huge dosnet with MANY zombies") ANYTHING can be taken down. For very large attacks of this type which exceed your own uplink bandwidth there are few options. However there is planning and preparation that can be done in advance to cope with events like this.
What i will say is this. We will never claim to be ddos proof as anyone with sufficient resources and intelligence can take anything down - but - much can be said for the network which is prepared, knowledgeable, and has failover plans in place. These are the things we have in our advantage.
We are also multi-homed & redundantly uplinked (3 x 1000Mb uplinks) with multiple physical locations, control our own network & backbone routers, and have several large IPv4 & IPv6 block assignments. (This pool is a hobby project... We actually own and operate an ISP as our mainline business).
So, yes. We have strategies in place. Are they effective? Only time can tell. We have to date been attacked once already which registered an additional 850Mb (megabit) of traffic on our upstream routers. This had no effect on us other than making some pretty spikes in our network utilization graphs BUT that is only one of several types of attacks that could be launched.
Cheers,
Anni
