Artificial Intelligence 5 min read

Implementation of a Cloud‑Native AI‑Powered Quantitative Research Platform Using Alibaba Cloud ACK

The article details how the Juqian intelligent investment research platform leverages Alibaba Cloud's ACK cloud‑native AI suite, Kubernetes, and various cloud services to build a high‑efficiency, scalable AI‑driven quantitative finance solution, improving resource utilization, reducing costs, and accelerating research workflows.

Alibaba Cloud Infrastructure
Alibaba Cloud Infrastructure
Alibaba Cloud Infrastructure
Implementation of a Cloud‑Native AI‑Powered Quantitative Research Platform Using Alibaba Cloud ACK

The 2024 Global Digital Economy Conference "Cloud·AI·Compute Ignite" was held in Beijing under the theme "Intelligent Cloud, Drawing Blueprint," showcasing the China Academy of Information and Communications Technology's research and practice in the AI cloud industry.

At the conference, Alibaba Cloud Container Service team, together with its customers, received the "2024 AI Cloud Native Typical Case" award.

As AI workloads become more central, traditional cloud services evolve into AI‑focused cloud offerings. The ACK Cloud‑Native AI Suite provides an open, extensible, high‑efficiency solution for AI/ML workloads on Kubernetes, and the awarded case – a cloud‑native AI‑based Juqian intelligent investment research platform – represents an early successful AI cloud deployment in quantitative finance.

The platform builds on Alibaba Cloud's ACK (Kubernetes edition) and integrates services such as NAS, OSS, SLS, GPU‑shared scheduling, HPA, and Prometheus to support factor calculation, model training, and strategy backtesting for quantitative research.

For compute, the platform combines fixed ECS nodes with elastic ECI nodes, forming a pool that raises average cluster utilization from below 40% to over 70%, delivering significant cost savings.

Storage is tiered: NAS handles frequently accessed, large, or numerous files, while OSS stores less‑frequent data, reducing I/O pressure and improving read/write performance.

Scheduling employs multiple components: basic tasks use Kubernetes orchestration; AI training/inference leverages GPU‑shared scheduling; data‑intensive tasks use ACK Fluid for unified NAS/OSS access and the Fluid Python SDK, boosting I/O performance and cutting compute costs to one‑tenth of the original.

By integrating AI Cloud, the platform shortens the end‑to‑end quantitative research cycle—from over a week for factor mining, model training, and validation to just one to two days.

The ACK Cloud‑Native AI Suite enhances productivity, efficiency, and cost‑effectiveness through resource sharing, efficient scheduling, elastic AI, data acceleration, and full‑lifecycle AI task management.

Alibaba Cloud's ACK suite has assisted users across internet, gaming, autonomous driving, quantitative finance, online education, AI for Science, and research institutions, supporting public, private, hybrid, and edge cloud deployments while fostering open‑source collaboration with the community.

cloud nativeAIKubernetesAlibaba CloudAI PlatformAckQuantitative Finance
Alibaba Cloud Infrastructure
Written by

Alibaba Cloud Infrastructure

For uninterrupted computing services

0 followers
Reader feedback

How this landed with the community

login Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.