Features: Dell PowerEdge RAID Controller (PERC) S100, PERC S300 User's Guide

Back to Contents Page


Dell PowerEdge RAID Controller (PERC) S100, PERC S300 User's Guide

  General Features


General Features

The features of the Dell PowerEdge RAID Controller (PERC) S100 and PERC S300 adapter are described in Table 3-1.

NOTE: BAS, BGI, CC and OCE run only on the operating system.

Table 3-1. Features of the PERC S100 controller and PERC S300 controller 

RAID Controller Feature


Automatic virtual disk rebuild

Rebuilds a redundant virtual disk automatically when a failure is detected, if a hot spare is assigned for this capability.

Background Array Scan (BAS)

Verifies and corrects correctable media errors on mirror, volume, or parity data for virtual disks. BAS starts automatically after a Virtual Disk is created while in the OS.

Background virtual disk initialization (BGI)

The background initialization of a redundant virtual disk creates the parity data that allows the virtual disk to maintain its redundant data and survive a physical disk failure.

Because background initialization helps the controller to identify and correct problems that might occur with the redundant data at a later time, background initialization is similar to a consistency check.

Background initialization allows a redundant virtual disk to be used immediately. Data is lost if a physical disk fails before the completion of a BGI.

NOTE: Although a BGI is software-initiated at the PERC Virtual Disk Management utility, the PERC S100 adapter/PERC S300 adapter drivers must be loaded before the BGI runs.

NOTE: Unless mentioned otherwise, the term PERC Virtual Disk Management utility refers to both the PERC S100 Virtual Disk Management utility and the PERC S300 Virtual Disk Management utility.

Boot support for RAID levels

Allows boot support for Volume, RAID 0, RAID 1, RAID 5, and RAID 10.

Boot support for degraded virtual disks

Enables the system to boot from degraded redundant virtual disks (RAID 1, RAID 5, or RAID 10).

Cache support for virtual disks

Supports these cache options: None, Read Only, Read/Write. The PERC S100 adapter or PERC S300 adapter uses part of system memory for cache.


Allows different types of checkpointing (background initialization, consistency check, and rebuild) to resume at the last point following a restart.

After the system restarts, background checkpointing resumes at its most-recent checkpoint.

Command queuing

Command queuing is a command protocol used by SATA and SAS physical disks that is supported by the PERC S100 adapter and PERC S300 adapter.

Command queuing allows the host to issue multiple input/output requests to a disk simultaneously. The disk can then decide in which order to process the commands to achieve maximum performance.

The SATA and SAS versions of command queuing have slightly different protocols and means of handling multiple traffic requests at the same time, but the end-purposes are comparable.

Consistency check (CC)

A consistency check is a background operation that verifies and corrects the mirror or parity data for fault-tolerant physical disks. It is recommended that you periodically run a consistency check on the physical disks.

By default, a consistency check corrects mirror or parity inconsistencies. After the data is corrected, the data on the primary physical disk in a mirror set is assumed to be the correct data and is written to the secondary physical disk in the mirror set.

A consistency check cannot be user-initiated in the PERC Virtual Disk Management utility. However, a consistency check can user-initiated when using Dell OpenManage Server Administrator Storage Management.

Disk initialization

For physical disks, initialization writes metadata to the physical disk, so that the controller can use the physical disk.

Fault tolerance

The following fault tolerance features are available with the PERC S100 adapter and PERC S300 adapter, in order to prevent data loss in case of a failed physical disk:

  • Physical disk failure detection (automatic).
  • Virtual disk rebuild using hot spares (automatic, if the hot spare is configured for this functionality).
  • Parity generation and checking (RAID 5 only).
  • Hot-swap manual replacement of a physical disk without rebooting the system (only for systems with a backplane that allows hot-swapping).

If one side of a RAID 1 (mirror) fails, data can be rebuilt by using the physical disk on the other side of the mirror.

If a physical disk in RAID 5 fails, parity data exists on the remaining physical disks, which can be used to restore the data to a new, replacement physical disk configured as a hot spare.

If a physical disk fails in RAID 10, the virtual disk remains functional and data is read from the surviving mirrored physical disk(s). A single disk failure in each mirrored set can be sustained, depending on how the mirrored set fails.

Mirror rebuilding

A broken mirror can be rebuilt after a new physical disk is inserted and the physical disk is designated as a hot spare. The system does not have to be rebooted.

Online Capacity Expansion (OCE)

OCE is a process that allows you to add storage capacity to an existing virtual disk. In most cases additional storage capacity can be added without taking the system offline. However, if an additional physical disk needs to be added and the system does not support hot-swapping, the system must be turned off.

OCE enables you to increase the total storage capacity of a virtual disk by integrating unused storage with the virtual disk.

Data can be accessed while the physical disks are added (if a system has hot-swap capability) and while data on the virtual disk is being redistributed.

For Volume and RAID 1, OCE expands the virtual disk by using the available space of the physical disks that are already members of the virtual disk. For RAID 0, RAID 5 and RAID 10, additional capacity can be attained by adding physical disks to the virtual disk.

Physical disks (general)

The PERC S100 adapter supports up to four SATA HDD or SSD physical disks. The PERC S300 adapter supports up to eight SAS or SATA HDD physical disks.

NOTE: The physical disks in a virtual disk must be the same interface and drive type (HDD or SSD). For example, you cannot mix a SATA and SAS interface (HDD or SSD), or HDD and SSD physical disks, in the same virtual disk.

A maximum of eight physical disks can be used for RAID 0 and RAID 5. A maximum of two physical disks can be used for RAID 1. A maximum of four physical disks can be used for RAID 10.

If a physical disk fails during system startup, the controller identifies the failed physical disk as follows:

  • At the PERC Virtual Disk Management utility by highlighting the failed physical disk in a virtual disk in red.


  • In a brief warning at the Dell Inc. PERC S100 Controller BIOS or Dell Inc. PERC S300 Controller BIOS screen, that a virtual disk(s) were found that are Degraded and/or Failed. This alerts the user to the failed physical disk(s).
  • At Dell OpenManage Server Administrator Storage Management.
  • With a bi-color Status LED on each physical disk. The green element of the Status LED is off, while the amber element flashes on and off.

NOTE: The Status LED applies only to systems with a PERC S300 adapter and a backplane that has removable physical disks.

NOTE: PERC S100 adapters support HDD physical disks of capacity 2 TB and greater. Current and upcoming releases of PERC S300 adapters will only support upto 2 TB HDD physical disks.

Physical disk hot-swapping
(hot-swap capability)

Hot-swap (hot-plug) capability is the manual substitution of a physical disk for another one while the host system is powered on.

If a system supports hot-swapping, physical disks can be plugged into a system's backplane while the controller is operating, without causing the controller to reset.

CAUTION: A physical disk can be hot-swapped from a system only if the system has a PERC S300 adapter and a backplane that supports hot-swapping.

NOTE: If a system's physical disks are accessible only when the system's cover is removed, the physical disks cannot be hot-swapped. The physical disks must be located on the backplane (behind the removable front panel) and accessible externally.

NOTE: When replacing physical disks in a virtual disk that has already been created, make sure that SAS HDD physical disks are replaced with SAS HDD physical disks, that SATA HDD physical disks are replaced with SATA HDD physical disks, and that SATA SSD physical disks are replaced with SATA SSD physical disks.

NOTE: When hot-swapping a physical disk, make sure that the new disk is of equal or greater capacity to the physical disk that is being replaced.

Physical disk roaming

The controller supports moving a physical disk from one backplane slot or cable connection to another (on the same controller). The controller automatically recognizes the repositioned physical disk and logically places it in the proper order.

A virtual disk is lost if you perform disk roaming while an OCE is running at Storage Management.

NOTE: For more information on how to open the system and add parts, see the Hardware Owner's Manual, available on the Dell Support website at

Perform the following steps for physical disk roaming:

NOTE: These steps do not apply to systems that have a backplane with hot-swap capability.

  1. Turn off the power to the system, physical disks, and system components.
  2. Disconnect the power cables from the system.
  3. Move the physical disks to different slots on the backplane or to different cable connections.
  4. Perform a safety check. Make sure the physical disks are inserted properly.
  5. Connect the power cables and power up the system.

NOTE: The controller detects the RAID configuration from the configuration data on the physical disks.

Storage port (Storport) driver support

For use with Microsoft Windows Server 2003 and later, the Storport driver improves throughput and miniport driver interfaces.

Stripe size

Stripe size is determined by a PERC S100 adapter/PERC S300 adapter algorithm. Stripe size cannot be configured by the user.

Virtual disks (general)


Up to eight virtual disks are supported.

The PERC S100 adapter and PERC S300 adapter allows:

  • Creating virtual disks of different RAID levels on a single controller.
  • Creating different RAID level virtual disks on the same physical disk, to adapt each virtual disk to the I/O that it processes.
  • Building different virtual disks with different characteristics for different applications.

The PERC S100 adapter and PERC S300 adapter does not allow:

  • Creating a virtual disk from a mix of different type physical disks. For example, a RAID 10 virtual disk cannot be created from two SATA-II HDD physical disks, a SAS HDD physical disk, and a SSD physical disk. All of the physical disks must be the same interface (SAS or SATA) and drive type (HDD or SSD).
  • Selecting a physical disk as a dedicated hot spare if the physical disk is a different type from the physical disk or disks.

A virtual disk refers to data storage created by the controller from one or more physical disks. Although a virtual disk can be created from several physical disks, it is seen by the operating system as a single disk.

The capacity of a virtual disk can be expanded online for any RAID level, without the operating system being rebooted.

Virtual disk migration

The controller supports automatic virtual disk migration from a PERC S100 adapter to a PERC S300 adapter (or vice versa). Manual intervention for migration is not required or used by the PERC S100 adapter or
PERC S300 adapter.

CAUTION: Before starting a virtual disk migration, power-down both systems before removing or inserting the physical disks. After the migration occurs, make sure that all of the physical disks have been migrated and are present in the virtual disk.

CAUTION: The virtual disk is lost if you perform a virtual disk migration during an OCE.

NOTE: The PERC S100 adapter only supports SATA HDD and SATA SSD disks. SAS disks cannot be migrated with a
PERC S100 adapter.

NOTE: A bootable virtual disk cannot be migrated between dissimilar controllers or dissimilar system models when the system uses Microsoft Windows Server 2003 as its operating system.

Virtual disk RAID levels

Virtual disks at different RAID levels can be created.

Virtual disk transformation


Virtual disk transformation can consist of:

  • Capacity expansion, using OCE (to allocate additional virtual disk space on the original physical disks or after additional physical disks are added).
  • Rebuilding (rebuilding data on a virtual disk consists of using an available hot spare or backup physical disk).


Table 3-2 compares the specifications of the PERC S100 adapter and PERC S300 adapter.

Table 3-2. Specifications for the PERC S100 adapter and PERC S300 adapter


PERC S100 adapter

PERC S300 adapter

SAS technology



SATA technology



eSATA technology



SSD technology



Support for x8 PCI-E Host Interface



I/O Controller

Intel ICH10R or
Intel Ibex Peak chipsets or Intel Cougar Point chipsets

Dell 3Gb/s SAS Adapter

Communication to the system


PCI-E lanes

Communication to end devices

SATA links

SAS/SATA links

SAS connectors


Two 4-port connectors on all systems

SATA connectors

Discrete on the motherboard

Two 4-port connectors on all systems




Supported operating systems

Microsoft Windows Server 2003 Family, Microsoft Windows Server 2008 Family,
Microsoft Windows Server 2008 R2

Dell-compliant SATA compatibility



Dell-compliant SAS compatibility



Dell-supported direct-connected end devices

Dell-compliant physical disks

Dell-compliant physical disks

*SMART error support through management applications



Backplane supported systems



Software-based RAID

Volume, RAID 1,

Volume, RAID 1,

Maximum number of virtual disks



Support for internal tape drive



Support for global hot spare



Maximum number of hot spares

Varies (by the number of free disks in the system)

Varies (by the number of free disks in the system)

*SMART is supported under the Windows Driver but is not supported with Unified Extensible Firmware Interface (UEFI) and CTRL-R. The SWRAID SMART drive status shows as "Degraded" in OpenManage Storage Services (OMSS).

Back to Contents Page