The following Isilon Q&A were compiled from “Ask the Expert: all about Isilon scale-out NAS storage array (Chinese Version)”. The original thread is https://community.emc.com/thread/184036.

Question 1:

What are the advantages of Isilon scale-out NAS storage compared with the traditional NAS storage? What are the commonly deployed ways of Isilon? Could you introduce the data layout of Isilon?

Answer:

The biggest advantage of Isilon is that OneFS works exclusively with the Isilon scale-out NAS nodes, referred to as a “cluster”. A single Isilon cluster consists of multiple “nodes”, which are constructed as rack-mountable enterprise appliances containing: memory, CPU, networking, non-Volatile Random Access Memory (NVRAM), low-latency InfiniBand interconnects, disk controllers and storage media. Each node in the distributed cluster thus has compute or processing capability as well as storage or capacity capabilities.

The Isilon deployment is very flexible. clients with adequate credentials and privileges can create, modify, and read data using one of the standard supported methods for communicating with the cluster. NFS, SMB, FTP, HTTP, HDFS and REST API are supported by Isilon. For example, Isilon are able to share the file system to Windows client via the SMB protocol, and meanwhile Isilon can also share the storage space to ESX as Datastore via NFS protocol.

OneFS is a completely symmetric and highly distributed file system. Data and metadata are always redundant across multiple hardware devices. Data is protected using erasure coding across the nodes in the cluster. For example, An Isilon cluster has 6 nodes, the data is spread across all six nodes by default. However, you are able to specify the data to store to the specific nodes or disks by setting diskpool. For more information about the Isilon system overview, please refer to the white paper “EMC ISILON OneFS: A Technical Overview”.

---

Question 2:

The new feature of Access Zone is supported in the OneFS 7.0, could you please introduce the concept of Access Zone?

Answer:

Access zone provide a method to logically partition cluster access and allocate resources to self-contained units, thereby providing a shared tenant environment. For instance, the customer can specify two spate Active Directory servers to different Zone for authenticating the SMB client in the OneFS 7.0 operation system. 

---

Question 3:

Does software update affect the original data in the Isilon? Is the cluster able to continue serving data to clients during a software upgrade?

Answer:

Cluster can be upgraded using two methods: Simultaneous or Rolling Upgrade. A simultaneous upgrade from 7.0.2 to 7.1 will involve a short period of data unavailability as the nodes reboot, but the process should be quick and painless. Rolling upgrades are only available in OneFS between versions whose third or fourth digits differ. Thus, 7.0.2.1 supports rolling upgrade to 7.0.2.4, or to any eventually 7.0.3 or 7.0.4 version.

If it is a small version update, you can select a rolling upgrade, restart the upgrade process nodes one by one, and the data is available. NFS data access will failover from the restarting node to other nodes. If it is a large version of the update, you need to select the simultaneous upgrade, the upgrade process nodes simultaneously reboots, and the client could not access data during the reboot, and need to reconnect to Isilon after the restart.

---

Question 4:

Each additional Zone will automatically increase a Local Provider; each Zone has a default User (Guest) and Group (Administrators, Users, Isilon Users, etc.). Do the default user and group exist in the separate zone? Could you please introduce the concept of role and the relationship between user and group?

Answer:

By default, Guest Users and Administrators groups have the same UID and GID in the various Access Zone. It is used for assigning permission to Windows default user and default group when using SMB sharing. The new feature of Role in the OneFS 7.0 is the role management tool, you can assign the different permission for Role, and then the user is given a Role. 

---

Question 5:

Could you please introduce the mechanism of Journal in Isilon?

Answer:

A file system journal, which stores information about changes to the file system, is designed to enable fast, consistent recoveries after system failures or crashes, such as power loss. The file system replays the journal entries after a node or cluster recovers from a power loss or other outage. Isilon uses two batteries to keep NVRAM in the powered state. When any one battery failure, the node will enter the read-only status to ensure the data safely.

Isilon does not rely on hardware RAID for protection of data, but it will protect files via Reed-Solomon error correction (N+M protection). Because all data, metadata, and parity information is distributed across the nodes of the cluster, the Isilon cluster doesn’t require a dedicated parity node or drive, or dedicated device or set of devices to manage metadata. The protection level by default at +2:1 provides for a situation where we wish to have double-disk redundancy and single-node redundancy. Of course, you can sacrifice some storage space to enhance the protection level.

---

Question 6:

Does Isilon have similar QoS features? Does Isilon also support the feature of multi-tenant?

Answer:

Currently, Isilon doesn’t support similar QoS features or multi-tenant feature.

---

Question 7:

Are we able to simultaneously use snapshots and remote replication in Isilon?

Answer:

Yes, Isilon supports to simultaneously use snapshots and remote replication.

---

Question 8:

Does tiered storage support automatic setting and manual setting? For example, user can set a directory or file to the high performance level.

Answer:

SmartPools allows multiple tiers of Isilon storage nodes to exist within a single file system. You can define the value of the data within your workflows based on Policies, and automatically aligns data to the appropriate performance tier over time. With automatic policies or manual control, you can tune performance and layout, storage tier alignment, and protection setting.

                                                                                                           

             

iEMC APJ

Please click here for for all contents shared by us.