manualshive.com logo in svg

Dell EqualLogic PS4100E, Configuration Manual

Share
Download
Dell EqualLogic PS4100E Configuration Manual Download Page 41

 

March 2013 

Dell EqualLogic Configuration Guide v14.1 

4-32

 

4.2

 

RAID 10 drive layouts and total reported usable storage 

Using a RAID 10 policy, Table 12 shows the drive layouts that are enforced based on the number of 
drives in each array/hot spare configuration, and the total usable storage available for each model. 

RAID 10 (mirrored sets in a striped set) combines two high performance RAID types: RAID 0 and RAID 
1. A RAID 10 is created by first building a series of two disk RAID 1 mirrored sets, and then distributing 
data over those mirrors. RAID 10 implementations can tolerate one drive failure per mirror pair. 

Table 12

 

RAID 10 drive layouts and total reported usable storage available with hot spares (in GB) 

Disk drives 

Hot spare 

No hot spare 

6 

4 Data (2 mirrors) + 2 Hot-spares 

6 Data (3 mirrors) + 0 Hot-spare 

7 

6 Data (3 mirrors) + 1 Hot-spare 

N/A

(a)

 

8 

6 Data (3 mirrors) + 2 Hot-spares 

8 data (4 mirrors) 

12

(c)

 

10 Data (5 mirrors) + 2 Hot-spares 

12 data (7 mirrors) 

14 

12 Data (6 mirrors) + 2 Hot-spares 

14 data (7 mirrors) 

16 

14 Data (7 mirrors) + 2 Hot-spares 

16 data (8 mirrors) 

24

(c)

 

22 Data (11 mirrors) + 2 Hot-spares 

24 data (12 mirrors) 

48 

46 Data (23 mirrors) + 2 Hot-spares 

N/A

(b)

 

(a)

 

Not supported. An even number of drives is required by RAID 10 

(b)

 

48 drive arrays cannot implement a no hot-spare policy. 

(c)

 

PS4100/PS6100 models only 

Total reported usable storage when using hot spares: All models prior to PS4100/PS6100 

Disk 

drive 

qty / 

Size 

50

(a)

 

100

(a) 

74 

146 

250 

300  400  450 

500 

600 

750  1000  2000  3000 

7

(b)

 

- 

- 

- 

- 

666 

- 

- 

- 

- 

- 

- 

- 

- 

- 

8

(c)

 

- 

- 

- 

- 

666 

- 

- 

- 

- 

- 

- 

- 

- 

- 

14

(d)

 

- 

- 

395 

- 

1333 

- 

2135 

- 

2671 

- 

4013 

- 

- 

- 

16 

309 

621 

- 

913 

1556 

1867 

2492 

2805 

3117 

3744 

4683 

6247 

12749  19230 

48 

- 

- 

- 

- 

- 

- 

- 

- 

10647  12779 

- 

21306  40840  63191 

Total reported usable storage when using hot spares: PS41x0/PS61x0

 

Disk 

drive 

qty / 

Size 

146 

200

(e) 

300 

400

(e)

 

500 

600 

900 

1000 

2000 

3000 

6 

- 

- 

- 

- 

887 

1064 

- 

1781 

3635 

5498 

12 

650 

920 

1331 

1843 

2222 

2672 

4014 

4454 

9103 

13762 

14 

781 

- 

1599 

- 

2670 

3207 

4816 

5353 

- 

- 

24 

1433 

2027 

2938 

4065 

4904 

5888 

8841 

9820 

19998 

30221 

(a)

 

Solid State Disk 

(b)

 

Legacy PS50E 

(c)

 

PS4000E and PS6000E only 

(d)

 

Legacy PS70E,PS100E, PS100X, PS200E and PS300E 

(e)

 

PS6100S (SSD) only 

Summary of Contents for EqualLogic PS4100E

Page 1: ...olutions based on EqualLogic PS Series arrays and the FS Series Family of NAS Appliances Recommendations and best practices for iSCSI SAN and scale out NAS network fabric design Updated capacity guidelines capabilities limitations and feature sets for the EqualLogic product family Version 14 1 March 2013 ...

Page 2: ...individual components switches network interface controllers initiators etc that make up the SAN infrastructure This configuration guide is intended for storage administrators SAN designers storage consultants or anyone who is considering purchasing or has purchased EqualLogic PS Series Arrays for use in a production storage area network THIS WHITE PAPER IS FOR INFORMATIONAL PURPOSES ONLY AND MAY ...

Page 3: ...3 3 September 2012 Updates Updated for PS M4110 Array Updated Blade Integration section to include Dell Force10 MXL Updated for FS7600 FS7610 and Fluid FS 2 0 Updated Controller Firmware Upgrade Path to Firmware 6 x Updated Section 2 4 RAID Policies Corrected table 25 in Section 4 7 1 New sections Section 4 Capacity Planning Section 6 2 Synchronous Replicaton SyncRep Section 6 3 Internet Protocol ...

Page 4: ...12 1 March 2011 Section 3 3 7 Replication Section 4 6 Integrating 1GbE and 10GbE SANs Section 2 1 Supported Configuration Limits maintaining consistency with current firmware release notes Appendix B and Appendix C Content removed and now published as a separate Validated Components List document All significant changes included in the most recent version of this document are highlighted using blu...

Page 5: ...ort failover behavior in PS4100 PS6100 controllers 1 12 1 4 5 Vertical port failover behavior in PS4110 PS6110 controllers 1 15 1 5 Array model PS M4110 1 17 1 5 1 Controller type in PS M4110 model 1 17 1 5 2 Configuration options 1 18 1 5 3 Failure behavior in the PS M4110 1 18 1 5 4 Networking considerations and guidelines 1 18 2 Controller firmware 2 20 2 1 About member firmware 2 20 2 2 Firmwa...

Page 6: ... total reported usable storage 4 33 4 4 RAID 5 drive layouts and total reported usable storage 4 34 4 5 Array RAID configurations and associated RAID sets 4 35 5 PS Series array concepts 5 36 5 1 Groups and pools 5 36 5 1 1 Pools 5 36 5 2 Volumes 5 38 5 2 1 Volume attributes 5 38 5 2 2 Volume features 5 39 5 3 Snapshots and clones 5 39 5 3 1 Clones 5 40 5 4 Thin provisioning 5 40 5 4 1 Template vo...

Page 7: ...ion 6 57 6 3 4 About IPsec Security Parameters 6 57 6 3 5 About IPsec Certificates 6 57 6 3 6 About IPsec Pre Shared Keys 6 58 6 3 7 About IPsec policies 6 58 6 3 8 IPsec considerations and limitations 6 58 6 3 9 Performance considerations 6 59 6 3 10 Host Connectivity Considerations 6 59 7 EqualLogic SAN design 7 60 7 1 General requirements 7 60 7 1 1 Implementation of standards 7 60 7 1 2 Genera...

Page 8: ...external switches for array connection 9 96 10 Fluid File system 10 103 10 1 FS Series architecture 10 104 10 1 1 FS Series solution for file only storage 10 104 10 2 Dell FluidFS 10 105 11 FS Series NAS Appliances 11 107 11 1 Equallogic NAS appliance supported configuration limits 11 107 11 2 Initial NAS cluster valid configurations 11 109 11 3 FS7500 system components 11 110 11 4 FS7500 file sys...

Page 9: ...31 14 5 Tested SAN designs 14 131 14 5 1 Blade IOM switch only 14 132 14 5 2 ToR switch only 14 132 14 5 3 Blade IOM switch with ToR switch 14 133 14 6 Data Center In A Chassis DCB design 14 134 14 7 VLANs for iSCSI 14 135 Appendix A Network ports and protocols 14 137 A 1 Required ports and protocols 14 137 A 2 Optional ports and protocols 14 137 Appendix B Recommended switches 14 139 Appendix C S...

Page 10: ...PS Series SAN and FS Series NAS appliance performance Note This version of the EqualLogic Configuration Guide v13 3 September 2012 introduces coverage of the EqualLogic PS M4110 Blade Array Unless otherwise stated the recommendations involving specific 10GbE solutions apply to the first generation PS6010 the second generation PS4110 PS M4110 and the PS6110 Unless otherwise stated recommendations i...

Page 11: ...cations and functions of PS Series arrays 1 1 Array models The array models available prior to introduction of the PS4100 PS6100 family are shown in Table 1 below Table 1 Array Models prior to PS4100 PS6100 Array Model Drive Type Number of Drives PS50E PS2400E SATA 14 16 a PS3000X PS3x00XV SAS 16 PS4000E SATA 16 PS4000X PS4000XV SAS 16 PS5000E SATA 16 PS5000X PS5000XV SAS 16 PS5500E SATA 48 PS6000...

Page 12: ...PS6110E 3 5 SAS 7 2K RPM 24 PS6110X 2 5 SAS 10K RPM 24 PS6110XV 2 5 SAS 15K RPM 24 PS6110XV 3 5 SAS 15K RPM 24 PS6110S SSD 12 or 24 PS6110XS SSD SAS 10K RPM 7 SSD 17 SAS PS M4110E 7 2K RPM NL SAS 2 5 14 PS M4110X 10K RPM SAS 2 5 14 PS M4110XV 15K RPM SAS 2 5 14 PS M4110XS SSD 10K RPM 2 5 SAS 5 SSD 9 SAS 1 2 PS Series supported configuration limits The Dell EqualLogic PS6xxx Series provides the ful...

Page 13: ...ximum of two PS4000 PS4100 and or PS M4110 arrays b Includes groups of mixed array types and all other group types except groups containing only PS40x0 PS41x0 and PS M4110 arrays When a group contains a mix of PS40x0 PS41x0 and PS M4110 arrays and other array models the higher limits prevail c Practical maximum volume size is operating system specific A PS Series group can create and present volum...

Page 14: ...ince the EqualLogic PS Series was introduced there have been several different array models released with new features better performance and greater storage capacity The storage array controllers were also improved to take advantage of advances in the underlying networking and storage technologies ...

Page 15: ...e type Notes Type 1 3 x 1GbaseT 3 x 1Gb SFP combo SATA Original Controller Design PS50 PS2400 1GB Cache Type 2 3 x 1GbaseT 3 x 1Gb SFP combo SATA PS50 PS2400 1GB Cache Type 3 SAS Type 3 SATA 3 x 1GbaseT SAS SATA PS3000 PS5000 1GB Cache Cannot mix Type 3 SAS with Type 3 SATA Type 4 3 x 1GbaseT SAS PS3000 PS5000 1GB Cache Cannot mix Type 3 SAS with Type 4 controller Type 5 3 x 1GbaseT SATA PS3000 PS...

Page 16: ...ssive controller cache mirrors the cache of the active controller 1 3 3 Controller failover behavior in all models prior to PS4100 PS6100 To support redundant controller fail over each Ethernet port on the active controller that is connected to the SAN must have its corresponding port on the passive controller also connected to the same SAN network In the event of a controller failure the passive ...

Page 17: ...rt 0 on the passive controller for the other connection to the SAN This is illustrated in the partially connected scenario shown in Figure 1 Figure 1 Partially connected controller failover Note how IP addresses are reassigned to the ports during the failover processes shown in Figure 1 and Figure 2 ...

Page 18: ...March 2013 Dell EqualLogic Configuration Guide v14 1 1 9 Figure 2 Fully connected controller failover ...

Page 19: ... x 1GbaseT 1 x 10 100Mb mgt SAS NL SAS SSD PS6100 Only 4GB Cache Supports vertical port failover Cache to Flash memory destaging Type 12 2 x 1GbaseT 1 x 10 100Mb mgt SAS NL SAS PS4100 Only 4GB Cache Supports vertical port failover Cache to Flash memory destaging Type 14 1 x 10GbE SFP 1 x 10GbaseT 10 100Mb mgt SAS NL SAS SSD PS6110 Only 4GB Cache Support for vertical port failover Cache to Flash me...

Page 20: ...Vertical port pairing insures that IP addresses assigned to each of the failed controller Ethernet ports apply to the corresponding ports on the second controller As stated in Section 1 3 3 above controller models prior to the PS4100 PS6100 required having cabled connections present on corresponding controller ports for controller failover to work correctly The vertical port failover feature in PS...

Page 21: ...e mode by the controller Thus a failure of a link or an entire switch would reduce bandwidth available from the array Referring to Figure 4 or Figure 5 assume that CM0 is the active controller In vertical port failover if CM0 senses a link drop on the local ETH0 port connection path it will automatically begin using the ETH0 port on the backup controller CM1 Vertical port failover is bi directiona...

Page 22: ...airs into a single logical port from the point of view of the active controller In a fully redundant SAN configuration you must configure the connections as shown in Figure 24 and Figure 25 in section 7 3 4 In a redundant switch SAN configuration to optimize the system response in the event you have a vertical port failover you must split the vertical port pair connections between both SAN switche...

Page 23: ...4 Figure 6 PS4100 vertical port failover and optimal connection paths IMPORTANT By alternating switch connection paths between ports in a vertical port pair port failover allows the array to maintain 100 bandwidth capability in the event of a switch failure ...

Page 24: ...from the array Referring to Figure 4 or Figure 5 assume that CM0 is the active controller In vertical port failover if CM0 senses a link drop on the local ETH0 port connection path it will automatically begin using the ETH0 port on the backup controller CM1 Vertical port failover is bi directional If CM1 is the active controller then vertical port failover will occur from CM1 ports to CM0 ports if...

Page 25: ...ch SAN configuration to optimize the system response in the event you have a vertical port failover you must connect either the SFP cable or the 10GbaseT cable from each controller to a different switch in the SAN network While it is a supported configuration it is not recommended to connect both the SFP and 10GbaseT ports on each controller at the same time In this scenario the SFP port will alwa...

Page 26: ...ed I O module connecting to the M1000e chassis infrastructure Controller Type Storage blade image Network Interfaces Storage type Notes Type 13 2 x 10Gb Ports One Per Controller Connected through the Backplane No Label Each of the two ports has an active link and a standby link to the fabric switches in the backplane of the M1000e chassis SAS NL SAS Dual hot pluggable 10GbE controllers 4GB of memo...

Page 27: ...ured with dual controllers active and passive connections and redundant fabrics to provide redundancy and prevent volume connections between hosts and SAN from being dropped in the event of a failure There are three failure scenarios that are protected by the array In the event of a link failure the active link of the active port goes down the standby link for the same port will take over In the e...

Page 28: ...ing the PS M4110 Storage Blade IOMs must be Interconnected Datacenter in a Box IOMs directly interconnected External Switches can be used to provide interconnection if rack mounted arrays are needed Must use Single Fabric I O module Placement Supported Fabrics are A and B only Fabric A not supported on older M1000e systems Mid plane v1 0 Mid plane upgrade kit available Fabric B supported on all M1...

Page 29: ...equallogic com support Each EqualLogic PS Series array runs a core operating system in firmware that provides all of the PS Series features and functionality The firmware version is defined using a version number and will be updated from time to time as new features are added or for general maintenance improvements The firmware version number takes the following form X Y Z X is used to identify th...

Page 30: ...ght to v5 0 4 If the array is already running v5 0 or v5 0 1 then you must first update to v5 0 2 before updating to v5 0 4 Table 6 Controller firmware upgrade paths Starting firmware version Ending firmware version 5 2 x Later 5 2 x releases 6 0 x 5 1 x Later 5 1 x releases 5 2 x 5 0 x excluding 5 0 0 and 5 0 1 Later 5 0 x releases 5 1 x 5 2 x 5 0 0 5 0 1 5 0 2 4 3 x Later 4 3 x releases 5 0 x ex...

Page 31: ...or Microsoft 4 0 x 3 5 1 4 0 x 3 5 x 4 0 x 3 5 x 4 0 x 3 5 x 3 4 x 3 3 x 4 0 x 3 5 x 3 4 x 3 3 x Host Integration Tools for VMware 3 1 2 3 1 x 3 0 x 3 1 x 3 0 x 2 0 x 3 1 x 3 0 x 2 0 x 2 0 x EqualLogic Storage Replication Adapter for VMware Site Recovery Manager 2 1 x 1 0 6 2 1 x 2 0 x 1 0 x 2 1 x 2 0 x 1 0 x 2 1 x 2 0 x 1 0 x 1 0 x EqualLogic Multipathing Extension Module for VMware vSphere 1 1 1...

Page 32: ... for High Availability and Non Disruptive Operations can prevent connections to volumes dropping application timeout group and or array failure 2 3 1 Verify hosts are designed for high availability Designing and configuring hosts for high availability helps minimize the chance of disruptions Recommended settings and configuration options on hosts running in the EqualLogic environment are detailed ...

Page 33: ...anges on the Linux hosts either at the time of installation or through the built in eqltune utility 2 3 6 Network topology health check Design your network for redundancy according to recommended best practices For all members arrays in a given SAN Group all ports should be connected to the same subnet Use at least two iSCSI SAN ports per host connected so that at least two of the host ports conne...

Page 34: ...drives according to the designated RAID level and assigns the appropriate number of spare drives The storage in the member is available after you set the RAID policy although array performance is not optimal until the system has completed verification of the new RAID configuration The Remote Setup Wizard specifies the RAID policy for members when it creates a group and adds the first member to it ...

Page 35: ... Line SAS hard drives that offer greater capacities than Class 1 If you do not know what types of drives are used in your array see the Dell EqualLogic PS Series Storage Arrays Release Notes which lists the types of drives used with each PS Series model In addition there is labeling on each drive to indicate its capacity and type 3 2 3 Spare drive policy Spare drives increase availability by provi...

Page 36: ... the only available RAID policy It is not a user selectable option RAID 50 Applications requiring an optimal combination of performance availability and capacity Systems containing fewer than 24 drives Systems containing 10K and 15K RPM drives RAID 5 Dell recommends against using RAID 5 for any business critical data although it may be required for certain applications depending on performance and...

Page 37: ...le below shows the supported RAID policy conversions The first column lists RAID configurations and the second column lists supported RAID policy conversions for them Table 10 Supported RAID policy conversions Current RAID policy Supported conversion RAID 10 All RAID 50 RAID 5 RAID 6 RAID 5 None RAID 6 None RAID 6 Accelerated None If a RAID policy conversion is not supported you can remove the mem...

Page 38: ...from a no spares RAID policy to a policy that uses spare drives you must use the CLI Refer to the Dell EqualLogic Group Manager CLI Reference Guide You can also use the CLI to convert to a RAID policy that does not use spare drives but Dell recommends against doing this ...

Page 39: ...ty for parity blocks Each parity block generates parity using a different view of the data blocks depending on the RAID 6 implementation RAID 6 can tolerate up to two drive failures per RAID stripe set at the same time without data loss RAID 6 is not recommended for workloads consisting mainly of random writes Using a RAID 6 policy Table 11 shows the drive layouts that are enforced based on the nu...

Page 40: ... the RAID set is reconstructed using the 450GB SAS hot spare e One 600GB hot spare 2x 600 GB parity and 2x 400 GB SSD parity In the event of a drive failure SSD or SAS the RAID set is reconstructed using the 600 GB SAS hot spare f PS4100 PS6100 models only Total reported usable storage when using hot spares All models prior to PS4100 PS6100 Drive Qty Size 50 a 100 a 74 146 250 300 400 450 500 600 ...

Page 41: ...Hot spares 12 data 7 mirrors 14 12 Data 6 mirrors 2 Hot spares 14 data 7 mirrors 16 14 Data 7 mirrors 2 Hot spares 16 data 8 mirrors 24 c 22 Data 11 mirrors 2 Hot spares 24 data 12 mirrors 48 46 Data 23 mirrors 2 Hot spares N A b a Not supported An even number of drives is required by RAID 10 b 48 drive arrays cannot implement a no hot spare policy c PS4100 PS6100 models only Total reported usable...

Page 42: ... Parity 2 Hot spares 24 data parity 48 b 46 Data Parity 2 Hot spares N A c a RAID 50 requires an even number of disk drives A 7 drive configuration without hot spare would result in odd number of disk drives b 48 drive arrays implement stripes across multiple RAID 5 sets within a single chassis c 48 drive arrays cannot implement a no hot spare policy d PS4100 PS6100 models only Total reported usab...

Page 43: ...or RAID 5 without a hot spare is not supported 7 6 Data Parity 1 Hot spare 8 7 Data Parity 1 Hot spare 12 a 11 Data Parity 1 Hot spare 14 13 Data Parity 1 Hot spare 16 15 Data Parity 1 Hot spare 24 a b 23 Data Parity 1 Hot spare 48 b 46 Data Parity 2 Hot spares a PS4100 PS6100 models only b 24 and 48 drive arrays implement multiple RAID 5 sets within a chassis Total reported usable storage when us...

Page 44: ...ID 6 1 Spare Disk 10 2 9 2 Yes RAID 10 2 Spare Disks 6 6 5 5 Yes RAID 50 2 Spare Disks 5 1 5 1 4 1 4 1 For selected configurations RAID 5 1 Spare Disk 12 1 9 1 Not for business critical data Table 16 shows the RAID set relationship for each RAID type in a 48 drive configuration Table 16 EqualLogic PS Series array RAID types and RAID set relationships Raid policy Spare disks Raid set relationship B...

Page 45: ...e information 5 1 1 Pools A pool is a container that each member is assigned after being added to the group A pool can have between one to eight members There is always at least one pool in any group and it is called the default pool unless the name is changed Regardless of the name of this pool it is always considered the default storage pool All newly added members are automatically assigned to ...

Page 46: ...red pools in a PS Series group The following best practices should be considered for storage pools unless the automatic performance load balancer APLB is disabled Mixing arrays of different drive speeds and types is fully supported and is the primary reason APLB was created Do not mix arrays with different controller speeds 1GbE 10GbE within a single pool unless they are each running unique RAID p...

Page 47: ... the create volume function in Group Manager or through the Command Line Interface CLI Each volume must have a unique name that is used by Group Manager to identify and manage the volume Volume names may be changed via Group Manager at any time Volume names must meet the following requirements 1 to 64 alpha numeric characters A Z a z 0 9 are legal characters Volumes are assigned a unique iSCSI Qua...

Page 48: ...n or after volume is created to store the snapshot s created from that volume All snapshot reserve space for a specific volume always resides in the same storage pool as the volume By default Group Manager allocates snapshot reserve space equal to 100 of the host volume s allocated space This ensures that a 100 data change in the volume can be protected by a single snapshot This value can be set t...

Page 49: ... Clone from Volume Creates a new volume Creates a new volume name Creates a new iSCSI target IQN Has same reported size as source volume Has same contents as source volume Cloned volume is immediately available Cloning consumes free pool space equivalent to 100 of the volume reserve setting for the source volume a Clone from Snapshot Clone from Inbound Replica a secondary group a Replica clones ar...

Page 50: ...he remaining available free space in the pool Set each thin provisioned volume s In use Volume Reserve Warning Level to a level that allows the administrator to ensure that additional physical storage can be deployed before the volume uses the remainder of current pool free space Ensure that the Sum of all In use Volume Reserve Warning Levels for all thin provisioned volumes does not exceed the cu...

Page 51: ... network path Table 19 provides definitions of EqualLogic PS Series terminology related to replication processing Table 19 Replication terminology Term Description Replica A time synchronized copy of an EqualLogic volume stored in a Secondary Group Replica Set A collection of all time synchronized replicas for a specific source volume Primary Group A group containing the source volume s to be copi...

Page 52: ...2 volumes configured for replication 6 1 2 Replication paths The example replication paths shown in Figure 12 are described below Basic partnership one partner group hosts the primary copy of the volume and a second partner group hosts the replica copy of the same volume We also show the reverse direction of the path for the Fast Failback replication feature if it is enabled for the replica set Si...

Page 53: ... 13 Replication Setup one time Configure replication partnership and volume replication settings Replication Processing repeating The primary group checks for availability of sufficient delegated and replica reserve space on the secondary group at the beginning of each replication processing phase If adequate space is not available then the process will pause and generate an event message Replicat...

Page 54: ...March 2013 Dell EqualLogic Configuration Guide v14 1 6 45 Figure 13 Replication Process ...

Page 55: ...ation Depending on the size of the volume the failback scenario can take significantly longer to complete if Fast Failback is not enabled 6 1 5 Sizing Replica Reserve and Delegated Space The volume replication process in EqualLogic PS Series storage consumes extra storage space on both the primary and secondary group partners In addition to the space consumed by the volume itself each volume will ...

Page 56: ...etained on the secondary site If a failback snapshot is retained on the primary group The default values that appear in Group Manager are sufficient to ensure that enough space is reserved for at least one successful replication even if the entire contents of a volume are altered between replicas Initial replication of a volume will cause the complete contents of the volume to be copied to the sec...

Page 57: ...n the Group Manager Help documentation b Start with 105 then add to that the maximum number of replicas expected to be stored in the replica set minus 1 multiplied by the expected Change Rate 6 1 6 Effect of TCP Window Size on Latency across WAN links In certain cases increasing the TCP Window Size setting on PS Series controllers can improve replication throughput across WAN links The ability to ...

Page 58: ...ume Warning Failure to provide access control to a shared volume can result in data loss By default EqualLogic PS Series groups disable multi host shared access to target If needed you can enable multi host access for a volume If you enable multi host access to a volume then proper I O operation concurrency must be maintained 6 2 Synchronous replication 6 2 1 About Synchronous replication Synchron...

Page 59: ...y to the SyncActive pool and the two pools are out of sync The group tracks all volume writes while SyncRep is paused and when the administrator resumes SyncRep writes the tracked changes to the SyncAlternate pool Out of Sync SyncActive pool and SyncAlternate pool may not contain the same volume data the SyncActive pool contains the most recent volume data A volume can become out of sync if SyncRe...

Page 60: ...member 6 2 6 How SyncRep protects volume availability In normal SyncRep operation in which the volume is in sync the pools containing the SyncActive and SyncAlternate volumes contain identical data Volume writes are accepted as shown in Fig 15 Figure 15 SyncRep 1 The iSCSI initiator sends a write to the group 2 Data is simultaneously written to both the SyncActive and SyncAlternate volumes 3 The S...

Page 61: ...ked or when tracked changes are being written back to the SyncAlternate volume 6 2 8 Tracked changes written to the SyncAlternate volume 1 The SyncAlternate volume becomes available 2 Changes tracked while the SyncAlternate volume was unavailable are written to it Until all tracked changes are written the data in the SyncAlternate volume is valid up to the point in time when the volume went out of...

Page 62: ...yncAlternate volume by following one of the procedures listed below Volume In Sync If the volume is in sync you may switch to the SyncAlternate as documented in the online help Although host access to the volume is disrupted during the switch no initiator changes are required Volume Out of Sync If the volume is out of sync the group administrator performs the steps outlined in the online help to s...

Page 63: ...plication writes bad data to the volume the bad data is simultaneously written to both the SyncActive and SyncAlternate pools Recovery Time If a disaster occurs in the primary group you can promote the replica set on the secondary group to a recovery volume After the promotion you must reconfigure initiators to discover and log in to the iSCSI target now hosted by the secondary group or switch to ...

Page 64: ... on the primary group every time a replica is created This allows for fast failback capabilities In addition you can schedule the creation of snapshots or create them on demand just as you would with any other volume See About SyncRep and Snapshots on page 126 for more information SyncRep creates snapshots of the volume whenever the SyncActive and SyncAlternate pools are switched Scheduling Replic...

Page 65: ...at least one member One PS Series group containing two storage pools each of which must contain at least one member 6 3 Protecting your EqualLogic group with Internet Protocol Security Internet Protocol Security IPsec is a set of standardized protocols designed to allow systems on IP based networks to verify each other s identities and create secured communication links IPsec uses cryptographic se...

Page 66: ...parameters control the authentication and key negotiation carried out using the Internet Key Exchange IKEv1 or IKEv2 protocol Security parameters specify the following Using IKEv1 IKEv2 or manual keying You can configure IPsec to use manual keys However manual keying provides significantly weaker security than IKEv1 or IKEv2 and is also significantly more difficult to configure Consequently Dell s...

Page 67: ...is allowed to pass If you want to drop all traffic that is not explicitly protected or passed you must create an IPsec policy that drops traffic by default If there are multiple IPsec policies in place the system determines their priority by the order in which they were created policies created first take precedence over policies created later You can also use IPsec policies to determine what traf...

Page 68: ... the group 6 3 10 Host Connectivity Considerations Enabling or disabling IPsec for the group using the IPsec enable and IPsec disable commands might disrupt host connectivity to the group for several minutes To prevent unplanned outages IPsec should be enabled or disabled during a planned maintenance window when there are no active iSCSI connections to any volumes Consult the documentation for you...

Page 69: ...ndards EqualLogic SANs are based on industry standards The following standards are required to support all host to target communications and member to member communications in an EqualLogic SAN IETF Standards o IETF RFC1122 Requirements for Internet Hosts Communications Layers o IETF RFC1123 Requirements for Internet Hosts Application and Support o IETF RFC3270 Internet Small Computer Systems Inte...

Page 70: ...ler 4x 1GbE ports total PS4110 family 1x 10GbE port per controller 2x 10GbE ports total PS5x00 family 3x 1GbE ports per controller 6x 1GbE ports total PS6x00 family 4x 1GbE ports per controller 8x 1GbE ports total PS6510 family 2x 10GbE ports per controller 4x 10GbE ports total PS6110 family 1x 10GbE port per controller 2x 10GbE ports total FS7500 NAS 12x1GbE ports per controller node four client ...

Page 71: ...ffic The actual VLAN configuration of your iSCSI SAN will be dictated by your SAN network design requirements and the features of the iSCSI SAN switches being used Jumbo frames should be enabled If you choose to use jumbo frames then all nodes in the SAN fabric must have jumbo frames enabled For best performance and reliability we recommend that all interconnection paths between non stacking switc...

Page 72: ...n to supporting data transfers between the hosts and the SAN Dell EqualLogic arrays also use the SAN to support inter array communication and data load balancing For this reason the more buffer space per port that a switch can provide the better Due to the multitude of buffer implementations used by switch vendors Dell cannot provide definitive guidelines as to how much is enough Port buffers shou...

Page 73: ...ell recommends this strategy where networking resources are limited Otherwise it is recommended to use a physically separated network dedicated to iSCSI traffic that is not shared with other traffic If sharing the same physical networking infrastructure is required then use Data Center Bridging DCB for EqualLogic SAN Support for creating Link Aggregation Groups LAG For non stacking switches the ab...

Page 74: ... consisting of 1Gb arrays 10Gb Switches The stacking link bandwidth should be at least 40Gb s in each direction on each wire full duplex to provide adequate throughput to support an EqualLogic SAN consisting of 10Gb arrays or a mix of 1Gb and 10Gb arrays 7 2 1 2 Non Stacking Switches Non stacking switches do not have a dedicated switch interconnect In this case to create an interconnect between th...

Page 75: ...tary options should be considered They may be very useful in allowing the network administrator to create a more efficient multi switch layer 2 network infrastructure for a SAN Be aware that these proprietary solutions must support the transmission of IEEE 802 1x flow control and jumbo frames if used to properly support an EqualLogic SAN 7 2 1 3 Using a LAG to Connect Stacked Switches In some situ...

Page 76: ...dwidth for each array added 10GbE switches attached to 10GbE array controllers PS6010 PS6510 or PS4110 6110 Random Small Block Workloads 1 5 arrays 20 30Gb of IST bandwidth between each switch 6 arrays At least 6Gb of IST bandwidth per array between each switch PS4110 PS6110 Large Block Sequential Workloads 1 5 arrays 10Gb of bandwidth per active array controller port up to the aggregated maximum ...

Page 77: ...cation or subnet with those in another location or subnet Advantages Leverages Ethernet standard extensions Can be used to interconnect switches from different vendors Can use Link Aggregation Protocols LACP EtherChannel to pool multiple 1GbE or 10GbE links into a single logical link providing bandwidth and redundancy Concerns Most solutions limited to 8 port link aggregation group Spanning Tree P...

Page 78: ...N Note As a best practice we recommend using the same NIC devices in your server for all connection paths to the iSCSI SAN This will minimize the complexity of server configuration management For maximum performance ensure that the PCI express slot hosting the network controller has the same specifications as the network controller For example if the network controller has a PCIe x8 interface then...

Page 79: ...ost port requirements must be met to use MPIO with EqualLogic SANs At least two 2 Ethernet ports are required on each host The host operating system must have a supported MPIO driver or service available The ports used for MPIO cannot be teamed to other ports The ports must be the same speed The ports must be assigned IP addresses on the same subnet EqualLogic MPIO General Recommendations Follow t...

Page 80: ...ings Customized settings may be required depending on the supported application s Change the Subnets included field to include ONLY the subnet s dedicated to the SAN network infrastructure membersessions should be set to the number of network ports dedicated to SAN subnet maximum of 4 volumesessions should be left at the default value of 6 totalsessions should be left at the default value of 512 7...

Page 81: ...plementations will vary depending on your network infrastructure requirements If sharing physical switches with other non SAN traffic we assume all switches are VLAN capable 7 3 4 Redundant SAN configuration In a redundant iSCSI SAN each component of the SAN infrastructure has a redundant connection or path The following figures show example connection paths necessary to create a Redundant SAN Fig...

Page 82: ... Figure 24Redundant SAN Connection Paths PS4100 Figure 25 Redundant SAN Connection Paths PS6100 Note For a production environment the configuration examples shown above will protect your access to data These are the ONLY SAN configurations recommended by Dell ...

Page 83: ...ply to both PS3000 PS6000 family controllers and PS4100 PS610 family controllers They are not recommended for production deployment since they do not provide end to end redundant connection paths 7 3 5 1 Single array controller configurations Table 27 below shows configurations using a single array controller Table 27 Single controller array configurations Single NIC Single Switch Single Controlle...

Page 84: ... array controller Table 28 Dual controller array configurations Single NIC Single Switch Dual Controller Single NIC Dual Switch Dual Controller Dual NIC Single Switch Dual Controller 7 3 6 Minimum cabling scenarios PS4100 and PS6100 The vertical port failover feature described in Section 1 4 4 allows you to cable dual controller PS4100 and PS6100 family arrays for maximum I O bandwidth and control...

Page 85: ... and Figure 27 show how to connect these arrays to accomplish this Note The example configurations shown in this section only apply to the PS4100 and PS6100 family arrays and are recommended only when you do not have available SAN switch ports necessary to support fully cabled configurations Figure 26Minimum cabling scenario PS4100 ...

Page 86: ...March 2013 Dell EqualLogic Configuration Guide v14 1 7 77 Figure 27 Minimum cabling scenario PS6100 ...

Page 87: ...flexibility based on array I O performance The PS Series Group Manager will allow the SAN administrator to still manage both types of arrays within the same SAN group The ability to mix 1Gb and 10Gb arrays supports seamless operational coexistence during during migration to a 10Gb SAN 8 1 Design considerations To properly implement a mixed speed SAN you must pay close attention to the following de...

Page 88: ... go through the slower 1Gb switches Figure 28Mixed speed redundant SAN using split interconnect between 1Gb and 10Gb switches 8 1 1 Optimizing Rapid Spanning Tree Protocol behavior The LAG between the 10Gb switches in Figure 28 creates a loop in the network Rapid Spanning Tree Protocol RSTP will compensate for this by blocking paths as necessary The optimal spanning tree strategy for this network ...

Page 89: ...he following list summarizes the important SAN design considerations for integrating 10Gb EqualLogic arrays into existing 1Gb EqualLogic SANs When integrating 10Gb switches into your existing 1Gb environment how you interconnect the mixed speed switches split vs straight uplink does not have a significant impact on performance as long as the uplinks are sized appropriately to your workloads If you...

Page 90: ...able or better performance across almost all situations If you have predominately 10Gb initiators you should only access data and volumes residing on 10Gb arrays from those initiators You may see high latency and retransmit rates when 10Gb initiators connect to 1Gb targets When adding 10Gb arrays place them in separate pools from your 1Gb arrays ...

Page 91: ...n a given blade server as described in Table 29 Each blade server has a LAN on Motherboard capability that is mapped to the I O modules located in the Fabric A I O modules slots on the M1000e chassis and only supports 1Gb or 10Gb Ethernet networking depending on the blade server model In addition each blade server has two mezzanine sockets for adding additional networking options such as 1Gb or 10...

Page 92: ...ch of these three general SAN design strategies the user must make design decisions with respect to the type of interconnects to use between the blade I O modules and or the external stand alone switches Depending on the design recommendation stacking link aggregation or a combination of both types of interconnect may be used In Section 9 1 we will discuss strategies for directly attaching arrays ...

Page 93: ...available on blade I O modules Current M Series blade I O modules have limited numbers of externally accessible Ethernet ports For a 1GbE SAN solutions each array will require up to four ports per I O module and for a 10GbE SAN solution each array will require up to two ports per I O module Based on these maximum requirements the table below provides a breakdown on the number of externally accessi...

Page 94: ...t be part of the same Layer 2 Ethernet fabric if they will be hosting connections from EqualLogic arrays For this reason if arrays are going to be directly attached to one or more blade chassis then the I O modules within the Fabric that will be hosting the arrays must be interconnected Since the I O modules are not interconnected via the chassis midplane the only alternative is to use external po...

Page 95: ... on the blade I O module model you may use either stacking or link aggregation as described earlier in Table 30 Where stacking is supported this will most likely be the primary method of interconnecting the blade I O modules to create a single switch fabric SAN switches can be located in any of the three enclosure fabrics and it is not required that the blade I O modules be in the same enclosure f...

Page 96: ...vantages Concerns Consistent M1000e fabric management that adheres to M1000e recommended practices for blade IO configuration Reduces administration overhead All blade server Ethernet ports used for SAN reside on the same mezzanine card resulting in a potential single point of failure Upgrading switch FW will require scheduled downtime for SAN network ...

Page 97: ...ntime for SAN network As discussed earlier one of the concerns when creating a SAN that consists of a single stack is that a single stack does not provide for a highly available solution when attempting to perform switch maintenance such as switch firmware updates If there is a requirement that the SAN solution cannot be brought down for regularly scheduled maintenance then stacking may not be the...

Page 98: ...a potential single point of failure Spanning Tree must be considered if uplinking SAN to external switches or additional M1000e enclosures Table 35 Single Enclosure Link Aggregation Dual Fabric Dual fabric non stacked configuration Advantages Concerns Ensures that Ethernet SAN ports on each blade server will be distributed across two different mezzanine cards for a more highly available solution S...

Page 99: ...the guidelines described in Section 9 2 If reducing administrative overhead is the goal the Direct Attached design with a inter fabric stack is the best option with the storage direct attached design requiring fewer cables thatn the host pass through design Direct Attached Single Chassis Stacked and or LAG Figure 31 SAN modules in single enclosure fabric 9 1 1 SAN design for multiple M1000e enclos...

Page 100: ...hing EqualLogic arrays As just indicated stacking of M Series I O modules should be the primary method of building a multiple enclosure SAN with arrays directly attached to the M Series I O modules As with the single enclosure discussion stacking has some advantages and disadvantages from a network management perspective and network availability perspective Take this into consideration before deci...

Page 101: ...dvantages Concerns Simplified switch management High bandwidth interconnect Not highly available single stack solution cannot be maintained without scheduled downtime Additional scalability will require SAN network redesign and possible downtime Other notes Limited scalability due to potentially excessive hop counts and latency Recommend no more than 2 3 enclosures in this design ...

Page 102: ...n to remote data center Requires more cables than stacked solutions Other notes Should only use 10GbE links for this strategy to ensure adequate bandwidth between modules and enclosures Limited scalability due to potentially excessive hop counts and latency Recommend no more than 2 3 enclosures in this design If large SAN solutions are required the design just discussed may not be the adequate Whi...

Page 103: ...k from the blade IOM switches The remaining ports on each switch can accommodate the connection of multiple PS Series array members The host storage port ratio with the maximum number of array members is 1 1 The following diagram illustrates how to directly connect the two switches and how the two switches are connected by an ISL LAG Also note that the corresponding port on the passive controller ...

Page 104: ...of M1000e enclosure Less expensive that multi tiered switch solution Supports non dell branded switching as a single vendor solution No firmware to update on the pass through modules Complex cabling in a rack environment Other Notes Additional M1000e enclosures are added in similar fashion to diagram above Supported with stacked or non stacked external SAN infrastructures This strategy works for a...

Page 105: ...his discussion will start with that basic SAN reference architecture The difference will be the connection of each M1000e enclosure s M Series I O modules to at least two switches in the external SAN infrastructure using either Link Aggregation the primary method or Stacking interfaces if the external switches are stack compatible See Stack Compatibility for M Series I O modules Even if the M Seri...

Page 106: ... ability to scale arrays Supports ability to scale host Supports sharing SAN with non M1000e servers External stacked switches can be expanded by adding additional switches to stack as more ports are needed Can support different vendor switches in the external tier Must bring external switch stack down to apply switch firmware updates Must bring internal switch stack down to apply switch firmware ...

Page 107: ...sharing SAN with non M1000e servers Can support different vendor switches in the external tier Limited scalability of LAG bandwidth Spanning Tree may block some links if a loop is created All switches must be managed separately Other Notes Recommend using 10GbE or larger links within inter switch LAGs to provide better bandwidth scalability Care needs to be taken to ensure that addition of switche...

Page 108: ... enclosures Supports ability to scale arrays Supports ability to scale host Supports sharing SAN with non M1000e servers Ease of Administration Must bring external switch stack down to apply switch firmware updates Must bring internal switch stack down to apply switch firmware updates Switches must be stack compatible Other Notes Recommend using 10GbE or larger links within inter switch LAGs to pr...

Page 109: ...ity to scale arrays Supports ability to scale host Supports sharing SAN with non M1000e servers High Availability Limited scalability of LAG bandwidth Spanning Tree may block some links if a loop is created Other Notes Recommend using 10GbE or larger links within inter switch LAGs to provide better bandwidth scalability Care needs to be taken to ensure that addition of switches does not cause a sp...

Page 110: ...M1000e servers Fewer switches to manage since each stack is managed as a single switch Firmware updates on external stack requires scheduled downtime Limited scalability of LAG bandwidth Spanning Tree may block some links if a loop is created Other Notes If available use 10Gb links between M series I O modules and external switches M series I O module stacks do not need to be interconnected since ...

Page 111: ...th non M1000e servers Fewer switches to manage since each stack is managed as a single switch Firmware updates on external stack requires scheduled downtime Limited scalability of LAG bandwidth Spanning Tree may block some links if a loop is created Other Notes If available use 10Gb links between M series I O modules and external switches M series I O module stacks do not need to be interconnected...

Page 112: ...ion requires FS Series Firmware Version 2 0 on the EqualLogic FS Series NAS appliances NAS antivirus service for CIFS shares NAS clusters support virus scanning by using an antivirus server You can configure up to four antivirus servers to scan files stored in CIFS shares NFS is not supported The NAS cluster solution contains integration with industry standard ICAP enabled antivirus software to en...

Page 113: ... level storage The storage pool shows a single pool that contains a NAS Reserve managed by the NAS Appliance plus other volumes that are providing block I O storage targets not used by the FS NAS appliance To emphasize this flexibility we show a group of servers connected to the NAS Appliance through the LAN Switches acting as CIFS NFS clients and connecting directly to the iSCSI SAN switches for ...

Page 114: ...ing standard CIFS and NFS file access protocols and authentication methods Active Directory LDAP NIS Some of the key features provided in FluidFS include Single namespace FluidFS presents a single file system namespace up to 509TB usable capacity accessible via a virtual IP address regardless of cluster size High availability and reliability The FS Series architecture includes a high speed cluster...

Page 115: ...the operation of PS Series array based snapshots Note FS Series FluidFS snapshots and PS Series volume based snapshots function independently and have no impact on each other Please see the following whitepaper for more information on Dell FluidFS Dell Fluid File System http www dellstorage com WorkArea DownloadAsset aspx id 1578 NAS Replication NAS replication is point in time and uses file syste...

Page 116: ...stem 4 per controller node FS7600 2U NAS appliance with two active active NAS controllers 2 x Intel 5620 Quad Core processors per NAS controller 24GB memory per NAS controller Redundant hot pluggable power supples and fans Client LAN 8x1GbE per appliance 4 per controller node iSCSI SAN 8x1GbE per system 4 per controller node FS7610 2U NAS appliance with two active active NAS controllers 2 x Intel ...

Page 117: ... Minimum container size 20 MB 20 MB Maximum container size Available NAS Reserve Available NAS Reserve Snapshots in a cluster 10 000 10 000 Snapshot schedules for a container 64 64 Snapshot schedules in a cluster 1024 1024 Snapshots per container 512 512 NFS exports in a cluster 1024 1024 CIFS shares in a cluster 1024 1024 CIFS active client connections in cluster 1500 3000 Local users in a cluste...

Page 118: ...ntrollers The limits shown for files and directories containers NFS exports CIFS shares and container snap shots are the maximum supported limits The group will not prevent you from exceeding a limit and will not warn you when you reach or exceed a limit 11 2 Initial NAS cluster valid configurations The table below describes the configurations allowed during initial configuration Table 47 Initial ...

Page 119: ...em components in the initial release of the EqualLogic FS Series NAS appliance 4 the FS7500 consist of two controller node units and one backup power supply BPS shelf containing two redundant power supplies The system components and required power cabling paths are shown in Figure 34 Figure 34FS7500 system components and power cabling 4 For detailed information on racking and power supply cabling ...

Page 120: ...the cache to disk in the event a controller node experiences a loss of power 11 4 FS7500 file system operation on controller failover In the FS7500 system writes to one controller node cache are mirrored to the other controller node cache before the write is acknowledged and committed This prevents loss of any in flight write I O if a controller failure occurs If a controller node failure occurs t...

Page 121: ...March 2013 Dell EqualLogic Configuration Guide v14 1 11 112 11 6 FS7610 components Figure 36FS7610 NAS Appliance FS7610 hot pluggable NAS Controller 2 FS7610 hot pluggable NAS Controller 1 ...

Page 122: ...eserve The NAS reserve is the storage space allocated to the NAS Cluster from an EqualLogic PS Series storage pool The NAS reserve has the following properties It resides within a single EqualLogic storage pool Only one storage pool in the EqualLogic storage group can be used for allocating the NAS reserve It is formatted with the Dell FluidFS A fixed amount of storage space in the pool 512GB per ...

Page 123: ...P address You can create a single large container or you can create many separate containers You can apply different backup snapshot security and quota policies to each container Creating multiple containers gives you the flexibility to apply different management policies to different containers The number and size of the containers in a NAS cluster depends on the storage needs of your NAS clients...

Page 124: ...ult security style for NAS containers is NTFS Controls file access by Windows permissions in all protocols A client can change the permission and ownership by using the Windows Security tab With NTFS file security style all access permissions are controlled by the Windows administrator by using access control lists or share level permissions Mixed Supports both NTFS and UNIX security styles The pe...

Page 125: ...nsiderations Each NAS Container has its own snapshot policy The minimum unit a snapshot can act on is a NAS Container The maximum number of snapshots you can retain per container is 512 The maximum snapshots retained per FS appliance cluster is 10 000 When you create a snapshot all shares and exports within that container are included If a particular share or export will require frequent snapshots...

Page 126: ... container can have its own snapshot policies The snapshots can be recovered instantly by administrators and are also available to end users through the NAS protocols as read only views 12 5 2 Replication The FS76X0 and FS7500 using FluidFS series firmware 2 0 now support snapshot based asynchronous NAS replication to or from another FS7500 or FS76X0 system in a different EqualLogic Group at a pee...

Page 127: ...FS Series Firmware FluidFS Replication of NAS containers within the same EqualLogic group or NAS cluster is not supported by FS appliances One EqualLogic group can form replication partnerships with 16 other groups however one NAS container in a group can be replicated to only a single replication partner NAS replication operates at block level After the first complete data replication only change...

Page 128: ...d Write with an ability to demote However all writes are lost when container is demoted to resume replication from primary site Permanent Promote Promote Read Write in case primary site or container is not available and clients need to be failed over to secondary site All writes are preserved ...

Page 129: ...appliance is comprised of two peer system controller nodes In the required fully cabled configuration each controller node requires thirteen separate Ethernet cable connections Thus a single FS7500 appliance comprised of two controller nodes requires a total of 26 Ethernet connections four connecting to the client LAN switches and nine connecting to the iSCSI SAN fabric switches Figure 40 shows th...

Page 130: ...his connection pattern is shown in Figure 41 Sizing the iSCSI SAN Inter Switch Connection for FS7500 The inter switch connection sizing guidelines provided in Section 7 2 2 also apply to FS7500 SAN design The FS7500 mirrors the write cache between controller nodes To accomplish this all write operations are transmitted across the controller interconnect Thus it is very important that you follow co...

Page 131: ... The Dell EqualLogic NAS appliances require the following networks Client network Used for client access to the NFS exports and CIFS shares hosted by the NAS cluster SAN internal network Used for internal communication between the controllers and communication between the controllers and the EqualLogic PS Series SAN The SAN and Internal networks use the same set of switches ...

Page 132: ...123 It is recommended to keep the client and SAN side networks physically separate and deploy two switches on both sides to protect against a switch failure Figure 42FS7600 AND FS7610 networks See figures below for network connections Figure 43FS7600 network ...

Page 133: ...March 2013 Dell EqualLogic Configuration Guide v14 1 13 124 Figure 44FS7610 network ...

Page 134: ...0 to 10Gb Ethernet by replacing the network controllers or the NAS appliance 10Gb functionality is provided by FS7610 Network The two iDRAC IPMI ports on FS7500 are 100Mb The SAN switches must be able to support 100Mb in addition to 1Gbps speed Most 10Gb SFP switches including 8024F and F10 S48 do not support 100Mb It is best to connect FS7500 to 1 Gb switches and then uplink 1Gb switches to 10Gb ...

Page 135: ...EEE 802 1p is not supported for EqualLogic implementations In a shared network environment LAN and SAN traffic can impact each other and QoS may not solve this because of QoS implementation differences with vendors These different implementations include Number of Queues managed Relative priorities between queues Bandwidth reserved per queue Also traditional QoS lacks selective flow control for ea...

Page 136: ... DCBx Data Center Bridging Capability eXchange Support for iSCSI application protocol priority with DCBX also known as iSCSI TLV Note DCB uses 10Gb and higher Ethernet only I O convergence using DCB for Dell EqualLogic iSCSI storage is the future direction for Converged iSCSI in a lossless Ethernet environment Without DCB it is typical to dedicate a separate physical network infrastructure for SAN...

Page 137: ...such as server LAN on a single network infrastructure as shown in Figure 46 iSCSI can be converged with non storage based server LAN traffic allowing the network ports and the inter connecting links to carry multiple traffic types or protocols A DCB enabled converged Ethernet infrastructure includes the NIC CNA on the end devices servers and storage arrays along with the switching infrastructure ...

Page 138: ...DCB for iSCSI when this functionality is enabled If any device in the SAN does not support DCB then DCB needs to be disabled at the switches for the entire SAN Once all devices in the SAN are DCB compliant then DCB can be re enabled Switches and server CNAs NICs that are designated as DCB Supported in the EqualLogic Compatibility Matrix have been fully validated by Dell to ensure compatibility for...

Page 139: ...xpected functionality or may be functional without meeting the converged network objectives Note It is not recommended to share network infrastructure with DCB and Non DCB components For a complete list of components that support the DCB standards see the EqualLogic Compatibility Matrix Additionally there is a partner verified support program PVSP between Dell and VMWare This agreement is to suppo...

Page 140: ...ingle layer switching with rack servers Figure 47Single Layer Switching example with rack servers PowerConnect 8132F supports the auto up port role for learning DCBX configuration and the auto down port role for internally propagating to a downstream device Similary Force10 S4810 supports the auto upstream and auto downstream port roles 14 5 Tested SAN designs This section shows three different en...

Page 141: ...witches No ToR switches are required The switch interconnect can be a stack or a LAG and no uplink is required Figure 48Blade IOM switch only 14 5 2 ToR switch only Network ports of both the hosts and the storage are connected to external ToR switches 10 GbE pass through IOM switches are used in place of blade IOM switches in the M1000e blade chassis The switch interconnect can be a stack a LAG or...

Page 142: ...33 Figure 49ToR switch only 14 5 3 Blade IOM switch with ToR switch Host network ports are connected to the M1000e blade IOM switches and the storage network ports are connected to ToR switches The switch interconnect can be a stack a LAG or a VLTi and should ...

Page 143: ...tier to the ToR switch tier is also required Figure 50Blade IOM switch with ToR switch Two tier design 14 6 Data Center In A Chassis DCB design Find guidance for incorporating a data center in a chassis with the new Dell EqualLogic PS M4110 Blade Array Force10 Blade IO Modules and PowerEdge M Series blade servers to form a reliable stable and well performing full end to end DCB solution ...

Page 144: ...is encoded in the VLAN tag this information will be lost if an end device sends iSCSI traffic in a default or native VLAN and the switch receives it in a tagged or trunk port As a result DCB prioritization and traffic classification will be lost All devices in the iSCSI data path must have the same VLAN ID configured on the respective ports participating in the iSCSI network to ensure proper funct...

Page 145: ...B requirements for EqualLogic to ensure that DCB is properly enabled and or disabled across all devices and to assist with identifying and resolving basic DCB configuration issues in the EqualLogic SAN see the following white paper EqualLogic DCB Configuration Best Practices For a comprehensive understanding of integrating an M1000e Blade Chassis in a Full End to End Data Center Bridging environme...

Page 146: ...following sections A 1 Required ports and protocols Table 49 lists the ports and protocols required for operating an EqualLogic iSCSI SAN Table 49 Required ports and protocols Type Port Protocol Access iSCSI TCP 3260 iSCSI To the group IP address and all individual member IP addresses EqualLogic Internal UDP 161 SNMP Management operations TCP 9876 Internal iSCSI intra system control TCP 25555 Inte...

Page 147: ...roup IP address TCP 3002 GUI communication To group IP address TCP 3003 GUI communication encrypted To group IP address SNMP UDP 161 SNMP To and from group IP address Syslog UDP 514 Syslog From group IP address EqualLogic Diagnostics TCP 20 FTP Software update and diagnostic procedures to all individual member IP addresses TCP 25 SMTP E mail and diagnostic notifications from all individual member ...

Page 148: ...Guide v14 1 14 139 Appendix B Recommended switches The list of recommended switches is now maintained in a separate document EqualLogic Compatibility Matrix http en community dell com dell groups dtcmedia m mediagallery 19856862 download aspx ...

Page 149: ...v14 1 14 140 Appendix C Supported iSCSI initiators The list of supported iSCSI initiators is now maintained in a separate document EqualLogic Compatibility Matrix http en community dell com dell groups dtcmedia m mediagallery 19856862 download aspx ...

Page 150: ... tool PS5000 Yes PS6000 Dellstar Yes PS6010 Dellstar Yes cus kit tool PS6000 None Yes PS6010 Dellstar Yes cus kit tool PS5500 Yes PS6500 Dellstar Yes PS6510 Dellstar Yes cus kit tool PS6010 None Yes cus kit tool Currently shipping arrays Latest available conversion model 1Gb to 10Gb conversion availability Drive upgrades availability PS6500 None Yes PS6510 Dellstar Yes cus kit tool PS6510 None Yes...

Page 151: ...ontent http en community dell com techcenter storage w wiki 2660 equallogic technical content aspx Rapid EqualLogic Configuration Portal http en community dell com techcenter storage w wiki 3615 rapid equallogic configuration portal by sis aspx Switch Configuration Guides http en community dell com techcenter storage w wiki 4250 switch configuration guides by sis aspx Storage Infrastructure and So...

Page 152: ...THIS WHITE PAPER IS FOR INFORMATIONAL PURPOSES ONLY AND MAY CONTAIN TYPOGRAPHICAL ERRORS AND TECHNICAL INACCURACIES THE CONTENT IS PROVIDED AS IS WITHOUT EXPRESS OR IMPLIED WARRANTIES OF ANY KIND ...

Reviews:

No comments

Brands by name

Popular brands

Load more brands