Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Java at Alibaba [pdf]

jcp.org

Post reply on HN
Print view View original

11–20 of 44 posts

Re: Java at Alibaba [pdf]

#11

Post by tanilama »

sago wrote: post #8

Either I'm being very dense, or something doesn't quite make sense about these numbers. The slides say that on the busiest day of the year, they were serving around 175,000 transactions per second (126k of which, or maybe additionally, were payments). On the next slide it says they have 'millions of JVMs' running an 'insurmountable number' of requests. Is each JVM really taking 10s+ per transaction? Even assuming tho…

Each transaction is likely composed of hundreds of requests to various of subsystems, and each of them might have vastly different constraints. I don't think the metric of average transactions across all JVMs they are running is a meaningful metric.

Re: Java at Alibaba [pdf]

#12

Post by yegle »

Cyph0n wrote: post #4

Impressive stuff. They have basically forked their entire backend stack, starting from the OS and moving upwards. I am particularly interested in how much AliOS deviates from mainline Linux.

I worked for Alipay, a subsidiary of Alibaba, around 2010. At that time, the tech stack was lagging behind compare to most silicon valley companies. But it was 7 years ago so things might changed since then.

Re: Java at Alibaba [pdf]

#13

Post by jjirsa »

sago wrote: post #8

Either I'm being very dense, or something doesn't quite make sense about these numbers. The slides say that on the busiest day of the year, they were serving around 175,000 transactions per second (126k of which, or maybe additionally, were payments). On the next slide it says they have 'millions of JVMs' running an 'insurmountable number' of requests. Is each JVM really taking 10s+ per transaction? Even assuming tho…

There’s probably a 10:1 or 100:1 page view to transaction ratio, and a 10:1 or 100:1 api request to page view ratio

200k transactions is a crazy number - a normal second for amazon is probably a few thousand orders (4-5k). This is 20x higher.

Re: Java at Alibaba [pdf]

#14

Post by jjirsa »

jjirsa wrote: post #13
sago wrote: post #8

Either I'm being very dense, or something doesn't quite make sense about these numbers. The slides say that on the busiest day of the year, they were serving around 175,000 transactions per second (126k of which, or maybe additionally, were payments). On the next slide it says they have 'millions of JVMs' running an 'insurmountable number' of requests. Is each JVM really taking 10s+ per transaction? Even assuming tho…

There’s probably a 10:1 or 100:1 page view to transaction ratio, and a 10:1 or 100:1 api request to page view ratio 200k transactions is a crazy number - a normal second for amazon is probably a few thousand orders (4-5k). This is 20x higher.

This tweet suggests I’m probably off by quite a bit (overestimating amazons orders)

https://twitter.com/joelcomm/status/917229394212675584

That’s $3400/second in transactions for amazon, and certainly their average price/transaction is more than $1:1

Re: Java at Alibaba [pdf]

#15

Post by _old_dude_ »

There was a presentation with more info at last JVM Language Summit https://www.youtube.com/watch?v=X4tmr3nhZRg

The scary things is that they have implemented coroutine (generalized async/await) by copying the stack back and forth instead of putting the stack in the heap like in stackless python or in scheme.

Re: Java at Alibaba [pdf]

#17

Post by tuyguntn »

_old_dude_ wrote: post #15

There was a presentation with more info at last JVM Language Summit https://www.youtube.com/watch?v=X4tmr3nhZRg The scary things is that they have implemented coroutine (generalized async/await) by copying the stack back and forth instead of putting the stack in the heap like in stackless python or in scheme.

For non-experts in this field could you please explain why its scary if it works great for their use cases?

Re: Java at Alibaba [pdf]

#18

Post by rqs »

yegle wrote: post #12
Cyph0n wrote: post #4

Impressive stuff. They have basically forked their entire backend stack, starting from the OS and moving upwards. I am particularly interested in how much AliOS deviates from mainline Linux.

I worked for Alipay, a subsidiary of Alibaba, around 2010. At that time, the tech stack was lagging behind compare to most silicon valley companies. But it was 7 years ago so things might changed since then.

It's certainly evolving.

Alibaba (and as well as other large online companies) had a huge leap during recent the Mobile Age as mobile phones makes online services more accessible to the public.

And government also enforced (impliedly) mobile carriers (China Unicom, China Mobile and China Telecom) to fueling this trend.

All this makes companies like Alibaba become very rich.

Plus, Alibaba's sites like taobao.com needs to handle tremendous amount of traffic in an average day, and even heavier traffic during promotion events like Nov 11 (Double 11) Day (It's like Black Friday[0]).

They had motive and resource to improve their system, and they opened some of those improvements already on GitHub[1]

[0] https://en.wikipedia.org/wiki/Black_Friday_%28shopping%29 [1] https://github.com/alibaba

Re: Java at Alibaba [pdf]

#19

Post by titzer »

tuyguntn wrote: post #17
_old_dude_ wrote: post #15

There was a presentation with more info at last JVM Language Summit https://www.youtube.com/watch?v=X4tmr3nhZRg The scary things is that they have implemented coroutine (generalized async/await) by copying the stack back and forth instead of putting the stack in the heap like in stackless python or in scheme.

For non-experts in this field could you please explain why its scary if it works great for their use cases?

Putting it in the heap would generally be safer, as there are fewer special cases to worry about in the garbage collector. Making a copy of the stack and managing it "manually" probably means a lot of special scanning logic in the GC.

Re: Java at Alibaba [pdf]

#20

Post by peoplewindow »

sago wrote: post #8

Either I'm being very dense, or something doesn't quite make sense about these numbers. The slides say that on the busiest day of the year, they were serving around 175,000 transactions per second (126k of which, or maybe additionally, were payments). On the next slide it says they have 'millions of JVMs' running an 'insurmountable number' of requests. Is each JVM really taking 10s+ per transaction? Even assuming tho…

Alibaba does a lot more than sell stuff on single's day.
Post reply on HN