Scalability and parallelization. Optimize control and dataflow. Distributed transactional storage, and many more — everything your need to build HPC backend for your service.
We offer Two storage \ database solutions:
each time
$ acapella run root.lua snapshot created: $TEST_USER/cli-launcher/7C7D0463 6 fragments not found in fragment store uploading fragments: f3.py root.lua lib1/f1.lua lib1/f2.py lib2/f1.lua lib2/f2.py snapshot is ready CPVM 1.2.744 (ea670221a04d3d26b18b17770b06e55e83290282) using 'transactional' preset start fragment: $TEST_USER/cli-launcher/7C7D0463:mult_root.lua transaction started: f39726e195844944 transaction completed execution time: without overheads (worker): 1 ms total (worker): 1 ms total (node): 2 ms transaction timestamps: start: 15:51:25.579000 [2017-11-12] rnd: 15:51:25.581000 [2017-11-12] counters: total conflicts: 0 total restarts: 0 async calls: 4 sync calls: 0 TVM reads: 210 TVM writes: 30 TVM traffic: TVM bytes read: 650 B TVM bytes write: 570 B $
Acapella «Code Fragments» saved in «CodeBase» lets application developers spin up code on demand in response to events originating from anywhere. Treat all Acapella and third-party cloud services as building blocks, connect and extend them with code, and build applications that scale from zero to planet-scale—without provisioning or managing a single server.
CPVM can automatically to try parallel execution and perform detecting of LPF access conflicts, if success — code work faster, otherwise — cpvm to correct mistake by restart «subtransaction tree»
Создание/деплой сервисов через swagger (yaml/json/xml).
Мгновенный запуск или обновление сервиса одной командой:
1 | acapella run my-service.yml |
.
Изоляция всех исполняемых фрагментов кода. Уровни доверия к коду, проработка ограничений безопасности
Задание окружения
исполняемого кода с помощью Dockerfile
Профилирование кода и накладных расходов io
Автоматический выбор узла для исполнения с оптимальными характеристиками
Запуск неограниченного количество подзадач, участвующих в одной транзакции
Запуск одновременно до 1 000 000 фрагментов/node
Пропускная способность до 10 000 исполнений/сек/node
Before
Inconsistencies lead to potential monetary loss during reconciliation. Global synchronous replication of trades is not feasible.
After
Cost savings and a consistent, unified, global view.
Before
Regional reach with many systems glued together.
After
Global reach with lower latency and a consistent view.
Before
Each server or cluster is its own universe.
After
Consistent, global view delivers a unified experience.
Your development agility comes from building systems composed of small, independent units of functionality focused on doing one thing well.
«Code Fragments» (Cloud Functions) lets you build and deploy services at the level of a single function, not at the level of entire applications, containers, or VMs.
you can configure environment for code fragments define Dockerfile or its name, CPVM automaticaly build, deploy and scale images and perform isolation and fast code runing
No server management
Scales automaticaly
Pay only while your code runs
Run code in response to events
Open and famillar
Connect and extents cloud services
CPVM profile your solution in background
CPVM select parallelism strategy in optimal granularity
CPVM analize our code and give recomandations
