Cloud Native Big Data Meetup｜Data Computing Infrastructure in the Cloud Native Era

With the popularity of cloud-native infrastructure, how to smooth cloud-native big data systems has become an irreversible trend . On June 10, the Volcanic Engine Cloud Native Computing and Rare Earth Nuggets developer community will hold a technology meetup in Shanghai, inviting four ByteDance big data infrastructure R&D experts to share their experiences and insights in the practical application of data computing infrastructure to help developers better understand the advantages and application scenarios of each technology .

⏰Event time: June 10, 2023 (Saturday)

⛳️Venue : Conference Room 622, Building C, Caohejing Center, No. 1520, Gumei Road, Xuhui District, Shanghai

Scan code to sign up

wonderful agenda

" Exploration and Implementation of Flink Based on Cloud Native"

Tao Kelu｜Volcano Engine Cloud Native Computing R&D Engineer

With the popularity of cloud-native and Kubernetes infrastructure, how to smooth cloud-native big data systems has become an irreversible trend. This sharing mainly discusses some explorations and practices of Flink's cloud nativeization.

This sharing will be carried out from the following aspects:

Flink on K8s community solutions and pain points
Byte Flink's practice sharing in the process of cloud nativeization
Cloud-native integrated Flink data lake practice sharing

" Ray's Cloud Native Practice in Bytedance"

Wanxing Wang｜Senior R&D Engineer of ByteDance Infrastructure

Ray is a new-generation computing engine that has emerged in recent years. Compared with traditional computing engines, it has better programmability and heterogeneous resource support capabilities. It is more and more widely used in distributed computing, especially in the field of machine learning. Ray has a wide range of applications in ByteDance, such as graph computing, ML, etc., and provides elastic operation capabilities based on the cloud-native environment.

This sharing will be carried out from the following aspects:

Ray Introduction and Advantages
Application scenarios of Ray in ByteDance
Kuberay Helps Ray Cloud Native Deployment

"Storage Acceleration in Cloud Native Scenarios"

Jun Guo｜Head of Big Data File Storage Technology of Volcano Engine

In the cloud-native scenario, the computing architecture of big data and machine learning is developing toward the separation of storage and computing, elastic scaling, and flexible scheduling. However, the bandwidth, latency, and affinity capabilities of various storage services are difficult to adapt to computing requirements. This difference has also given birth to the development of the field of storage acceleration, and various storage acceleration services are currently available on the market. However, in the face of the diversity of computing and customer scenarios, there is currently no industry-standard storage acceleration practice, and many customers also face a lot of confusion when making model selections.

This sharing will focus on the practical summary of building storage acceleration capabilities:

What kind of storage acceleration capabilities do computing services need?
How to make storage acceleration easier and easier to use
Storage Acceleration Practices in Multiple Scenarios

"Technical Practice of Elasticsearch Project in Byte"

Lu Yuncheng｜ByteDance Infrastructure Senior R&D Engineer

The Elasticsearch project is currently the most popular distributed search and analysis engine. ES has large-scale cloud-native applications in ByteDance and provides very elastic service capabilities internally.

This sharing mainly introduces:

The Evolution of Byte under the ES Cloud Native Architecture
Kernel enhancements for ES
ES Ecological Capability Expansion

Cloud Native Big Data Meetup｜Data Computing Infrastructure in the Cloud Native Era

Guess you like