近幾年來，因為資料量成長的速度急劇上升，如何有效率的管理和存取這些資料變得越來越重要。對於這樣的情況來說，混合雲是一種適合的解決方法。在我們的系統中，我們一方面使用公共雲來存放所有的檔案，另一方面也將私有雲當作快取來使用。由於壓縮這項技術，可能會為混合雲這樣系統帶來更多優勢，因此我們將它應用在我們的系統之中。在這篇論文中，我們分別對公共雲和私有雲應用了以壓縮為基礎的方法來改善系統整體效能和減少在公共雲上對於儲存的花費。在私有雲上，我們會在存取檔案但失誤時，為該檔案以適合的壓縮方式壓縮後再存入私有雲；而在公共雲上，我們則為所有的檔案選擇適合他們的壓縮方式。當我們在做選擇時，我們除了會考慮每個檔案的特性（如檔案大小、讀和寫的比例）和每種壓縮演算法的特性（如壓縮比、壓縮和解壓縮時的計算時間）之外，也會考慮私有雲目前的狀態以及使用者對於公共雲反應時間的要求。最後我們也分別以模擬和真實世界的數據來評估我們的方法，而兩者的結果都說明在系統上應用壓縮確實能改善效能以及減少儲存方面的花費。

The amount of data is rapidly in recent years.
The ability to store those data and the efficiency of accessing become more important.
Hybrid cloud storage is a suitable solution for such a situation.
We use public cloud storage to backup all files, and use private cloud storage as cache in our system.
While compression have the chance to enhance the benefit from hybrid cloud storage, we apply it to our system.
In this paper, we introduce an approach for both public and private sides to improve performance and save storage cost.
We choose a suitable compression algorithm for the requested file when it is missing in private side.
And we find a combination of compression algorithms for all files in public side.
In our approach, we make the decision based on file characteristics(file size, write-read ratio, etc) and algorithm characteristics (compression time, decompression time and compression ratio).
Moreover, we also consider the cache list in private side and the constraint of response time in public cloud storage.
We evaluate our approach using simulation and real trace, and the improvement of performance and the saving of storage cost can be seen in the results.

1. Adaptive Cache Compression for High-Performance Processors
2.DEPSKY: Dependable and Secure Storage in a Cloud-of-Clouds
3.SCFS: A Shared Cloud-backed File System
4.Deduplication and Compression Techniques in Cloud Design
5.Measurement and Analysis of Large-Scale Network File System Workloads
6.Workload Analysis of a Large-Scale Key-Value Store
7.Hybris: Robust Hybrid Cloud Storage
8.To Zip or not to Zip: Effective Resource Usage for Real-Time Compression
9.Characteristics of File System Workloads
10.A Comparison of File System Workloads
11.Cmpres: To Save Your Cloud World Compressed
12.RACS: A Case for Cloud Storage Diversity
13.Frugal Storage for Cloud File System
14.Evaluating Adaptive Compression to Mitigate the Effects of Shared I/O in Clouds
15.Cumulus: Filesystem Backup to the Cloud
16.Efficiency of lossless data compression
17.An Approach for Data Privacy in Hybrid Cloud Environment
18.Adaptive Compressed Caching: Design and Implementation
19.Cost-Optimal Scheduling in Hybrid IaaS Clouds for Deadline Constrained Workloads
20.ECM: Effective Capacity Maximizer for High-Performance Compressed Caching
21.Comparison of Lossless Data Compression Algorithms for Text Data
22.C-Pack: A High-Performance Microprocessor Cache Compression Algorithm
23.A Hybrid Cloud Storage Architecture for Service Operational High Availability
24.SAM: A Semantic-Aware Multi-Tiered Source De-duplication Framework for Cloud Backup