Blizzard, Move to AWS (DDOS)

[Clannad-darkspear (Clannad)]

OP’s position is a typical “IT Manager” position. Apparently if you move everything to AWS, its going to be immune to computer problems.

If you don’t know what you’re doing then you’re 100% going to have the same issues. Tell me, when is the last time Amazon has been hit offline by a DDOS. Would love to know.

[Turbocharged-area-52 (Turbocharged)]

I take it you weren’t around in the mid 70s.

[Inni-thorium-brotherhood]

Ok, you’re partially right about Netflix.

[Clannad-darkspear (Clannad)]

And they account for 15% of internet traffic globally. When is the last time you’ve seen them taken down?

[Inni-thorium-brotherhood]

When is the last time you’ve seen them taken down?

A couple of weeks ago. That’s easy to find.

Date	Started	Ended	Duration
August 31 2019	18:40	18:57	00:17:51
August 31 2019	15:45	17:39	01:53:58
August 31 2019	11:55	15:12	03:17:54
August 07 2019	12:30	13:12	00:42:02
August 06 2019	16:00	16:02	00:02:04
July 30 2019	18:40	18:49	00:09:41
July 30 2019	13:20	13:26	00:06:37
July 25 2019	04:40	04:59	00:19:10
July 18 2019	16:00	16:13	00:12:59
July 17 2019	17:20	17:33	00:13:42
July 10 2019	12:20	12:45	00:25:06
July 09 2019	18:10	19:18	01:08:31
June 14 2019	02:50	03:51	01:01:33
June 06 2019	14:20	15:04	00:44:43
May 03 2019	20:00	20:17	00:17:38
May 03 2019	13:40	14:25	00:45:32
April 25 2019	13:10	14:59	01:49:00
April 04 2019	16:20	16:35	00:15:42

[Clannad-darkspear (Clannad)]

Ok, you’re partially right about Netflix.

As for other big companies:
Adobe, Airbnb, Alcatel-Lucent, AOL, Acquia, AdRoll, AEG, Alert Logic, Autodesk, Bitdefender, BMW, British Gas, Canon, Capital One, Channel 4, Chef, Citrix, Coinbase, Comcast, Coursera, Docker, Dow Jones, European Space Agency, Financial Times, FINRA, General Electric, GoSquared, Guardian News & Media, Harvard Medical School, Hearst Corporation, Hitachi, HTC, IMDb, International Centre for Radio Astronomy Research, International Civil Aviation Organization, ITV, iZettle, Johnson & Johnson, JustGiving, JWT, Kaplan, Kellogg’s, Lamborghini, Lonely Planet, Lyft, Made, McDonalds, NASA, NASDAQ OMX, National Rail Enquiries, National Trust, Netflix, News International, News UK, Nokia, Nordstrom, Novartis, Pfizer, Philips, Pinterest, Quantas, Sage, Samsung, SAP, Schneider Electric, Scribd, Securitas Direct, Siemens, Slack, Sony, SoundCloud, Spotify, Square Enix, Tata Motors, The Weather Company, Ticketmaster, Time Inc., Trainline, Ubisoft, UCAS, Unilever, US Department of State, USDA Food and Nutrition Service, UK Ministry of Justice, Vodafone Italy, WeTransfer, WIX, Xiaomi, Yelp, Zynga.

[Hypêrspace-dalaran (Hypêrspace)]

well i know who kermit the frog is. is that what you mean? haha. no idea what the joke is tho.

[Caîus-tichondrius (Caîus)]

ITT: Multiple sides who think they are smart and knowledgeable to the point they think the other shouldn’t even participate in the conversation…and everyone involved uses google as their primary and only knowledge base.

[Clannad-darkspear (Clannad)]

Was no google involved. It’s been common knowledge that Activision and companies like Netflix are heavily leveraging AWS. What’s actually in this thread is one side entirely making stuff up.

[Algolagnist-mankrik]

Oh my god, not this again.

If DDOS mitigation were so trivial, do you think it would still be such a huge problem? People are determined to blame the victim. Unreal.

[Clannad-darkspear (Clannad)]

It’s provided as a service on AWS. Have you ever seen Netflix or Amazon itself entirely taken down to a DDOS attack? It’s a trivial thing for them.

People in here just have no idea how something like AWS works. You can set up a virtual private cloud with instances load balanced across different availability zones. Amazon has it down to each AZ being on different power grids, ISPs, and even fault lines for earthquakes. Come on… And after all of that they provide DDOS protection for free as a service and their advanced version is miles beyond that which is something a company like Blizzard can, but probably don’t want to, pay for.

[Dãisy-emerald-dream]

You can educate my ignorance. The way I look at it services like netflix can be distributed, since the data can come from various locations and data centers, whereas blizzard has actual servers (realms) which isn’t duplicated information that isn’t distrubted, so when an IP is getting DDOS’d, blizzard has to change IP’s continuously from each attack. I feel like the design of a MMORPG makes what you are asking for more difficult in comparison to other games.

[Algolagnist-mankrik]

Lol facepalm.

[Clannad-darkspear (Clannad)]

You can educate my ignorance. The way I look at it services like netflix can be distributed, since the data can come from various locations and data centers, whereas blizzard has actual servers (realms) which isn’t duplicated information that isn’t distrubted, so when an IP is getting DDOS’d, blizzard has to change IP’s continuously from each attack. I feel like the design of a MMORPG makes what you are asking for more difficult in comparison to other games.

There is a lot of truth behind what you say. It would be much harder to handle the realms across WoW being distributed than a lot of other services or games. It’s not impossible, but definitely harder. In reality a game like WoW is stateful. Serverless architecture is generally stateless. Thing is, AWS provides anti-DDOS services by default and their paid versions are even better. Let AWS handle stuff like this and let Blizzard focus on the game is all I’m saying. When something like this happens the fault goes back to AWS and they are tasked with handling how to better modify their prevention instead of it being Blizzard themselves scratching their heads about what to do with their internal servers/networking.

[Fabermor-frostwolf]

I know where Netflix hosts many of their servers, I know people who have physically worked on them. I also know where Activision Blizzard has some servers, I have physically walked past them.

Straight up 10 IQ and arguing from authority with no substance. Activision was literally featured at AWS:ReInvent.

They’ve literally done a case study on Netflix on AWS that will take you half a second to google. You’re lucky I can’t post links.

Thats good for them? This isnt’ an issue that Autoscale can fix, and they have REGION lock their servers

[Ned-norgannon]

Channel 4

Well if Channel 4 does it, then dagnabit I’m convinced.

[Clannad-darkspear (Clannad)]

You don’t want autoscale to handle this anyways. I’m stating AWS has better DDOS protection, with AWS shield, than internal Blizzard stuff. Let Amazon handle that and focus on the game.

[Desarian-aggramar]

yes both netflix and amazon have both been taken down but its completely dishonest trying to use a website and a streaming service and compare it to the ddosing of a game. its 10x easier to ddos a game than it is a website furthermore its dishonest to try and use netflix and amazon because the whole point of ddosing is for attention. which the prick ddosing is gonna get far quicker bring down a game than they will going after a website.

[Fabermor-frostwolf]

You don’t want autoscale to handle this anyways. I’m stating AWS has better DDOS protection, with AWS shield, than internal Blizzard stuff. Let Amazon handle that and focus on the game.

Except it doesn’t… This has nothing to do with their level of protection, a DDOS attack out of the gate is about swamping a pipe… Literally trying to shove a turkey into a straw so nothing else can get in. AWS still has limits per region as to how much traffic they can/will handle… And for their SHield to work, the traffic has to get to them first… However if there’s still more traffic hitting the routes/gateway than the pipe can handle… you get the same problem

And AWS will be more likely to shut down a client’s VPC if they get hit often… Also they do not hold 90% of all Cloud customers

[Widdly-burning-legion]

After a bit of research I’ve come across the general consensus that Blizzard hosts their actual games on their own internal servers in their own cloud. I could be wrong here as Blizzard has been pretty quiet on how they operate, but it’s drastically clear they don’t leverage third party cloud providers outside of smaller things like their launcher. It’s clear because the standard is AWS (Amazon’s cloud) and anyone hosting on their services are blasted to the public by Amazon and generally the company leveraging their services openly acknowledge their product being hosted on AWS.

With that being said, I think it’s time to move towards AWS, Blizzard. Let them handle security, hardware updates, defunct hardware replacement, system administration costs, networking, etc, etc. AWS provides you with the standard for cloud services. It’s literally their job and their product. I can’t even imagine the manpower and costs behind designing, setting up, and maintaining your own internal cloud. And for those out of the loop I’m referring to actual server warehouses. What Blizzard is doing is creating a product, which is the game, alongside another product to actually host it. Who do you think provides better DDOS protection, security, reliability, etc, etc. A company devoted to games or a company devoted to cloud services?

This would have been shot down pretty quickly, outside of the ISP hits, if everything was hosted on AWS and set up properly and no one can tell me otherwise.

Notice: A few regions for OW are hosted on AWS. They are regions that come off as areas Blizzard wants no part in to set up server warehouses. The servers you connect to in the Americas are internal to Blizzard.

You would be surprised at the amount of people here who know nothing about databasing and servers that will tell you that this is wrong. Don’t listen to those people, they are idiots.

This is absolutely correct. Virtual distributed servers would mitigate the majority of these DDoS attacks.