CurveFS can manage different storages (HDFS, OSS, EBS) below • Apps access data by POSIX interface • Infrequent data is moved to OSS, and frequent data is moved to high speed storage transparentlyData long NETWORK LATENCY 50MS 1s frequently 7s recentlyStorage Engine Comparison (vs. Ceph) META MANAGEMENT CURVE CHUNK SERVER BLUESTORE META Precreate Chunk File Pool on ext4 RocksDB META OVERHEAD CurveFS • support operator capability level 2: automated application provisioning and configuration management and patch and minor version upgrads supported • File meta data preallocate • RAFT optimization

Supports RDMA and SPDK ○ Further reduce I/O latency and improve throughput ○ Support data lifecycle management ○ Curve block devices can be used as the underlying storage of Curve file systems ● System operation provide high availability/reliability ● Zero copy; Data stripe; RDMA to improve performance ● Management and monitor tools ● Support CSI deriver for upper cloud native applications accessTHANKS

CURVE COMMENT BASIC INSTALL Y (by Helm) automated application provisioning and configuration management SEAMLESS UPGRADES Y (by Helm) patch and minor version upgrads supported FULL LIFECYCLE Plan

{ OK = 0; // UNKNOWN_ERROR = 1; // NOSPACE = 2; // } // fs interface message GetFsInfoRequest { optional uint32 fsId = 1; // fs id optional string fsName DeleteFsInfoResponse { required MetaStatusCode statusCode = 1; } service MdsService { // fs interface rpc CreateFs(CreateFsRequest) returns (CreateFsResponse); rpc MountFs(MountFsRequest) returns OK = 0; // UNKNOWN_ERROR = 1; // NOSPACE = 2; // } // dentry interface© XXX Page 7 of 15 message GetDentryRequest { required uint32 fsId = 1; required uint64

and NBD use it to access storage services provided by Curve. The client provids posix-like interface, such as read/write/aioread/aiowrite in data plane and open/create/rename/extend, etc in control complete the disk write operation. CurveFS architecture Curvefs implements the file system interface based on FUSE. 1. Fs-meta Cluster is used to manage the inode and dentry metadata of files.

SpaceStatusCode status = 1; // required uint64 chunkId = 2; } service SpaceAllocService { // space interface rpc InitSpace(InitSpaceRequest) returns (InitSpaceResponse); rpc AllocateSpace(AllocateSpaceRequest)