4 ms·

Matrix of Services

28 points by espinchi 12y ago

cjbprime 12y ago
This is just describing Service-oriented Architecture (SOA) -- for a better read, see Steve Yegge's Google Platforms Rant: https://plus.google.com/+RipRowan/posts/eVeouesvaVX https://plus.google.com/+RipRowan/posts/eVeouesvaVX
- asinglenet 12y ago
  Yes, MAXOS is a variant of SOA. And, service architectures seem to have won the test of time. However, the modern version of service architecture has a very different management philosophy from the SOA of the last decade. The old SOA was to put components in place with a top down design. The new version is designed to evolve with lightweight and bendable protocols, peer-to-peer interactions, and continuous integration. The old version of SOA focused on defining the protocols and API's that services would use to communicate, from the top down. There was a lot of effort put into standardized object definitions and CORBA and EJB, and standardized protocols like SOAP. The resulting services were put into directories and locked down. That's what made SOA so annoying and cumbersome. It was a lot of work just to set up the plumbing, and then you can't change it to do what you want. I call the new version "service architecture" because "service oriented architecture" has such stifling connotations. This version uses a variety of protocols which are faster, lighter, and sometimes more specialized. It can include the old RPC protocols, new platform-specific RPC, connection pools, message queues, HTTP-REST, and plain old HTTP. It's not so important to define the protocols up front, because the whole system is always being tested as a unit. Reliability is being created by continuous integration, not by up front engineering and standards. There is little effort to package things into similar code objects. Services from many different languages and platforms co-exist. That's the whole idea. That said, I believe that services packaging may converge into something portable like Docker containers. As my friend Aaron O'Mullen of Codebox said, a Web service is the new executable.
  - dragonwriter 12y ago
    > The old version of SOA focused on defining the protocols and API's that services would use to communicate, from the top down. There was a lot of effort put into standardized object definitions and CORBA and EJB, and standardized protocols like SOAP. I think that CORBA and EJB are sort of the institutionalization of the SOA concept, in much the way that rigid, top-down practices like Scrum are part of the institutionalization of Agile. I don't think original SOA was any more heavyweight, top-down than original Agile; I think people tend to naturally gravitate toward centralized, top-down ways of doing things, producing a natural cycle of "back to basics", followed by centralizing, top-down approaches trying to impose more definition and manageability on the "new" approaches. (And are aided in this by tool vendors, who want people to buy exclusively into their tools rather than have heterogenous environments, for clear commercial reasons.) There's also new techniques that develop in parallel to the cycle and stop the cycle from strictly being reinventing the wheel, usually, but the basic pattern is largely cyclical. > Services from many different languages and platforms co-exist. That's the whole idea. That was a widely touted feature of old SOA, and of earlier (but by now somewhat old) reactions to the way old SOA went as heavily platform dependent services became common in it.
  - platz 12y ago
    > I call the new version "service architecture" because "service oriented architecture" has such stifling connotations. That's why the term 'microservices' is being pushed. It would be better to just agree on 1 new term instead of several new terms.
drtse4 12y ago
"Matrix of Services, which I will call MAXOS, is the secret weapon that Silicon Valley is using to disrupt and destroy competitors." I think i will stop here, thanks.
- AVTizzles 12y ago
  Not so secret?
  - drtse4 12y ago
    Yep, and too many useless hyperboles in one sentence.
- hackcasual 12y ago
  Followed shortly by: None of these companies uses traditional product management or textbook “Agile” methods Nope! Does the author think Agile is done company wide?
platz 12y ago
latency - how do you build a platform with high performance if every customer request has a network of n services to traverse for every operation?
- Denzel 12y ago
  You'll want to issue requests concurrently, use timeouts liberally, and implement fallbacks to cached/calculated data in the event that one of the services fails or misbehaves. Those are just a few guiding principles that will help when building a SOA (service-oriented architecture). Netflix has a great blog post on the structure of their platform [1]. They even open-sourced Hystrix [2], the library they use internally. [1] http://techblog.netflix.com/2012/02/fault-tolerance-in-high-volume.html http://techblog.netflix.com/2012/02/fault-tolerance-in-high-... [2] https://github.com/Netflix/Hystrix https://github.com/Netflix/Hystrix
- yureka 12y ago
  There are a lot of simple techniques that can be used to optimize services in a SOA (e.g. caching). Since each service follows a request/response format, it becomes easier to see where the majority of a request lifespan spends its time to help eliminate bottle necks. You'd be surprised how low the overall latency for a request can be in large companies like Amazon (< 100ms)
- jfoutz 12y ago
  what's your budget? Is 30ms fast enough to put a widget in a shopping cart? Let's just assert 15ms from your server to the customer, you've got 15ms to play with. Given this https://gist.github.com/jboner/2841832 https://gist.github.com/jboner/2841832 let's assume you get around 5ms for each service, say a couple big disk reads and some filtering. So if n=3, it doesn't seem to bad. How much can you do in parallel? Stuff like search, it seems like you could partition quite a bit hit a bunch of services (10?)in the first 5ms, then spend the remaining 10ms sorting results, just send your best effort. I'm (of course) assuming a customer request means there's a customer sitting there waiting and 30ms is good enough. Won't work for a game. Won't work for HFT. I guess you'd use the standard stuff, many layers of caching, efficient implementation of the specific services, good algorithms. Heck, you could play a game with docker and run the whole infrastructure on 1 big machine, i think those local sockets would be a lot quicker that the 500,000 ns of a datacenter read. I mean, if an l2 cache miss is a disaster for you, clearly this won't do anything for you. But how fast is fast enough? do you need to hit that target every single time(real time)? Can you be late 1 in a million requests? Are you sort of dismissing the idea out of hand or are you (aside from the silly title) really thinking through the consequences of implementing an application this way?
Fuzzwah 12y ago
I only skimmed the article but it made me think about Valve Software's cabal based development. http://www.gamasutra.com/view/feature/131815/the_cabal_valves_design_process_.php http://www.gamasutra.com/view/feature/131815/the_cabal_valve...

28 points by espinchi 12y ago

This is just describing Service-oriented Architecture (SOA) -- for a better read, see Steve Yegge's Google Platforms Rant: https://plus.google.com/+RipRowan/posts/eVeouesvaVX https://plus.google.com/+RipRowan/posts/eVeouesvaVX

asinglenet 12y ago
Yes, MAXOS is a variant of SOA. And, service architectures seem to have won the test of time. However, the modern version of service architecture has a very different management philosophy from the SOA of the last decade. The old SOA was to put components in place with a top down design. The new version is designed to evolve with lightweight and bendable protocols, peer-to-peer interactions, and continuous integration. The old version of SOA focused on defining the protocols and API's that services would use to communicate, from the top down. There was a lot of effort put into standardized object definitions and CORBA and EJB, and standardized protocols like SOAP. The resulting services were put into directories and locked down. That's what made SOA so annoying and cumbersome. It was a lot of work just to set up the plumbing, and then you can't change it to do what you want. I call the new version "service architecture" because "service oriented architecture" has such stifling connotations. This version uses a variety of protocols which are faster, lighter, and sometimes more specialized. It can include the old RPC protocols, new platform-specific RPC, connection pools, message queues, HTTP-REST, and plain old HTTP. It's not so important to define the protocols up front, because the whole system is always being tested as a unit. Reliability is being created by continuous integration, not by up front engineering and standards. There is little effort to package things into similar code objects. Services from many different languages and platforms co-exist. That's the whole idea. That said, I believe that services packaging may converge into something portable like Docker containers. As my friend Aaron O'Mullen of Codebox said, a Web service is the new executable.
- dragonwriter 12y ago
  > The old version of SOA focused on defining the protocols and API's that services would use to communicate, from the top down. There was a lot of effort put into standardized object definitions and CORBA and EJB, and standardized protocols like SOAP. I think that CORBA and EJB are sort of the institutionalization of the SOA concept, in much the way that rigid, top-down practices like Scrum are part of the institutionalization of Agile. I don't think original SOA was any more heavyweight, top-down than original Agile; I think people tend to naturally gravitate toward centralized, top-down ways of doing things, producing a natural cycle of "back to basics", followed by centralizing, top-down approaches trying to impose more definition and manageability on the "new" approaches. (And are aided in this by tool vendors, who want people to buy exclusively into their tools rather than have heterogenous environments, for clear commercial reasons.) There's also new techniques that develop in parallel to the cycle and stop the cycle from strictly being reinventing the wheel, usually, but the basic pattern is largely cyclical. > Services from many different languages and platforms co-exist. That's the whole idea. That was a widely touted feature of old SOA, and of earlier (but by now somewhat old) reactions to the way old SOA went as heavily platform dependent services became common in it.
- platz 12y ago
  > I call the new version "service architecture" because "service oriented architecture" has such stifling connotations. That's why the term 'microservices' is being pushed. It would be better to just agree on 1 new term instead of several new terms.

drtse4 12y ago

"Matrix of Services, which I will call MAXOS, is the secret weapon that Silicon Valley is using to disrupt and destroy competitors." I think i will stop here, thanks.

AVTizzles 12y ago
Not so secret?
- drtse4 12y ago
  Yep, and too many useless hyperboles in one sentence.
hackcasual 12y ago
Followed shortly by: None of these companies uses traditional product management or textbook “Agile” methods Nope! Does the author think Agile is done company wide?

platz 12y ago

latency - how do you build a platform with high performance if every customer request has a network of n services to traverse for every operation?

Denzel 12y ago
You'll want to issue requests concurrently, use timeouts liberally, and implement fallbacks to cached/calculated data in the event that one of the services fails or misbehaves. Those are just a few guiding principles that will help when building a SOA (service-oriented architecture). Netflix has a great blog post on the structure of their platform [1]. They even open-sourced Hystrix [2], the library they use internally. [1] http://techblog.netflix.com/2012/02/fault-tolerance-in-high-volume.html http://techblog.netflix.com/2012/02/fault-tolerance-in-high-... [2] https://github.com/Netflix/Hystrix https://github.com/Netflix/Hystrix
yureka 12y ago
There are a lot of simple techniques that can be used to optimize services in a SOA (e.g. caching). Since each service follows a request/response format, it becomes easier to see where the majority of a request lifespan spends its time to help eliminate bottle necks. You'd be surprised how low the overall latency for a request can be in large companies like Amazon (< 100ms)
jfoutz 12y ago
what's your budget? Is 30ms fast enough to put a widget in a shopping cart? Let's just assert 15ms from your server to the customer, you've got 15ms to play with. Given this https://gist.github.com/jboner/2841832 https://gist.github.com/jboner/2841832 let's assume you get around 5ms for each service, say a couple big disk reads and some filtering. So if n=3, it doesn't seem to bad. How much can you do in parallel? Stuff like search, it seems like you could partition quite a bit hit a bunch of services (10?)in the first 5ms, then spend the remaining 10ms sorting results, just send your best effort. I'm (of course) assuming a customer request means there's a customer sitting there waiting and 30ms is good enough. Won't work for a game. Won't work for HFT. I guess you'd use the standard stuff, many layers of caching, efficient implementation of the specific services, good algorithms. Heck, you could play a game with docker and run the whole infrastructure on 1 big machine, i think those local sockets would be a lot quicker that the 500,000 ns of a datacenter read. I mean, if an l2 cache miss is a disaster for you, clearly this won't do anything for you. But how fast is fast enough? do you need to hit that target every single time(real time)? Can you be late 1 in a million requests? Are you sort of dismissing the idea out of hand or are you (aside from the silly title) really thinking through the consequences of implementing an application this way?

Fuzzwah 12y ago

I only skimmed the article but it made me think about Valve Software's cabal based development. http://www.gamasutra.com/view/feature/131815/the_cabal_valves_design_process_.php http://www.gamasutra.com/view/feature/131815/the_cabal_valve...