Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[doc]add asr usercase #42

Merged
merged 1 commit into from
Nov 22, 2023
Merged

[doc]add asr usercase #42

merged 1 commit into from
Nov 22, 2023

Conversation

Cyber-SiKu
Copy link
Contributor

No description provided.

CurveFS 兼容两种存储后端的特性,使其可以适应不同业务的存储需求,既保证高性能,也控制存储成本。


### CurveFS 优势
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

优势改成应用价值

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

优势改成应用价值

fix


### CurveFS 优势

相较于本地存储而言,CurveFS 具有以下优势:
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这里是本地存储,下面是对比3副本存储,需要统一。


* 统一管理训练数据,支持多节点共享访问

* 后端接入 NOS低频存储,相比3副本存储每年每 PB 数据存储可节约40%成本,大幅降低 ASR 训练存储成本
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这里是3副本。上面是本地。我看他们原来是打算每个节点都放一份训练数据?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这里是3副本。上面是本地。我看他们原来是打算每个节点都放一份训练数据?

是的


* 后端接入 NOS低频存储,相比3副本存储每年每 PB 数据存储可节约40%成本,大幅降低 ASR 训练存储成本

* 支持数据多级缓存,提升训练并发效率
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

具体提升效果补充下

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

具体提升效果补充下

fix


作为一个年轻的项目,Curve仍在快速迭代。未来Curve会继续优化在AI和大数据分析场景的适配能力:

与各类AI框架深度集成,提供自动化的数据预热、训练优化等功能,提升AI训练的易用性和效率。
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

格式需要改善,预览下看看。

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

格式需要改善,预览下看看。

fix

@@ -0,0 +1,74 @@
# CurveFS 助力网易云商,解决 ASR 训练数据增长需求
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

题目建议不要使用ASR缩写 -- 直接用人工智能 语音识别之类的 这样更吸引人?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

语音识别

fix

随着语音技术在各行各业的广泛应用,自动语音识别(ASR)正在成为众多云服务商的核心竞争力。
但 ASR 模型的持续优化需要大量训练数据的支持,如何高效管理海量训练数据成为云商面临的一个难题。

网易云商在发展 ASR 业务过程中,存储面临了巨大挑战。
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

遇到的这几个困难可以按照 "无序列表 "模式展示

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

遇到的这几个困难可以按照 "无序列表 "模式展示

fix

CurveFS 兼容两种存储后端的特性,使其可以适应不同业务的存储需求,既保证高性能,也控制存储成本。


### CurveFS 优势
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这里可以参考下https://kms.netease.com/article/88340 , 内容更丰富一些

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这里可以参考下https://kms.netease.com/article/88340 , 内容更丰富一些

#43
已经使用了?

Signed-off-by: Cyber-SiKu <[email protected]>
@aspirer aspirer merged commit 16670fc into opencurve:main Nov 22, 2023
1 check passed
@Cyber-SiKu Cyber-SiKu deleted the doc/usercase branch November 22, 2023 02:45
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants