Storing Data in Windows Server 2019 PDF

Storing Data in Windows Server 2019 System Administration FWD 216 Topics Understanding storage technologies Understanding RAID Understanding disk types Exercise: enabling dedup on Windows Server 2019 Understanding storage technologies Besides high processing powe...

Storing Data in Windows Server 2019 System Administration FWD 216 Topics Understanding storage technologies Understanding RAID Understanding disk types Exercise: enabling dedup on Windows Server 2019 Understanding storage technologies Besides high processing power, a sufficient amount of RAM memory, and several network connections, your server will likely require a large amount of storage space too. Regardless of whether it is a single server or a cluster of servers, technologies such as IDE, SAS, SCSI, DAS, NAS, SAN, and RAID represent a wide range of storage technology options at your disposal. Different storage types Different types of storage technologies exist. Let's take a look at some of these devices: Optical discs: They offer large capacities and read-and-write accepted speeds. However, they continue to play the role backup media for data. HDDs: They offer large capacities and high read-and-write speeds. They continue to maintain the number one spot in the secondary storage category. SSDs: With their growing capacities and extraordinary read- and-write speeds, they are becoming a popular storage technology. Only time will tell if they can take the top spot from HDDs. ATA, PATA, SATA, and SCSI interfaces Advanced Technology Attachment (ATA), Parallel ATA (PATA), Serial ATA (SATA), Small Computer System Interface (SCSI) are the interfaces that are used to connect storage devices and peripherals to computers. ATA, also known as Integrated Drive Electronics (IDE), is a legacy interface that is used to connect HDDs, optical disc drives, floppy disk drives, and related storage technologies to computers.  ATA, PATA, SATA, and SCSI interfaces The two most popular types of ATA interfaces are as follows: PATA uses a 40-pin connector and cable for data transfer in order to connect the storage device to the computer's motherboard and uses Molex as a power connector to connect the storage device to the computer's power supply. The disk controller resides on a drive itself. SATA represents a replacement for the PATA interface and is widely used in personal computers. It uses a 7-pin cable for data transfer to connect the storage device to the computer's motherboard and a 15-pin power supply connector to connect the storage device to the computer's power supply. Like PATA, in SATA, the disk controller is located on a drive as well. ATA, PATA, SATA, and SCSI interfaces SCSI is another interface that connects storage devices and peripheral devices to computers. The two most popular types of SCSI are SPI and SAS. SPI is the early version of SCSI. SAS is the modern version of SCSI that provides high data transfer rates, and is widely used in servers. Peripheral Component Interconnect (PCI) and PCI Express (PCIe) In the mid-90s, PCI replaced IBM's Industry Standard Architecture (ISA), a 16-bit built-in expansion slot on the motherboard. Unlike the ISA, Intel's PCI has a 32-bit and 64-bit builtin slot on the motherboard that enables the expansion of computer capabilities. Later on, with the increase in demand for faster speeds, PCI was replaced by PCIe. PCIe is a serial expansion bus standard that comes with four connections: PCIe x1, PCIe x4, PCIE x8, and PCIe x16. It transmits data in full-duplex mode (sending and receiving at the same time) over wires known as lanes. Understanding local storage Local storage refers to the hard drive or SSD that's directly attached to the server. In the following subsection, we will talk more specifically about local storage. Direct-attached storage (DAS) DAS is a group of disks that are directly connected to computers or servers. You won't be wrong if you think of your computer's HDD as a DAS. However, in addition to internal storage devices, even external storage devices that are connected to computers or servers with any of the aforementioned interfaces are considered as DAS. Understanding network storage In contrast to local storage, network storage refers to the storage device that is connected to a computer network in order to provide data access to users. Network-attached storage (NAS) NAS is a network appliance that connects to computers and servers through a switch and acts as dedicated storage in an organization's network. There are manufacturers that build NAS as a file server. It brings flexibility so that organizations can rely completely on NAS for file sharing services without the need to use other servers. Storage area network (SAN) SAN, like DAS and NAS, is a storage technology too. However, unlike DAS and NAS, SAN is a standalone infrastructure. If we refer to the designation, then SAN is almost identical to LAN. While LAN connects computers on a local computer network, SAN connects storage devices in the area network. As such, a SAN cannot be accessed by other LAN devices. Just because of that, proprietary protocols or Simple Network Management Protocols (SNMPs) provide management for a SAN. Usually, Ethernet or FC is used to connect SAN to servers. Differentiating between block-level storage and file-level storage The comparison between file-level storage and block- level storage is shown in the following table: File-level storage Block-level storage Data is stored and accessed in the Data is stored in blocks that form of files and folders. represent volumes, which are which are then managed by the OS. Used by NAS. Used by SAN. Understanding adapter and controller types The disk controller is an electronic circuit that resides on a hard disk. It performs operations such as spinning disks, moving heads for reading and writing, and transferring data to and from RAM. Unlike the disk controller, the Host Bus Adapter (HBA) controller is an electronic board that is integrated with or attached to the motherboard of the server so that it can connect the storage system to the server. Understanding serial bus technologies For data transmission, parallel and serial transmissions are used. In parallel transmissions, a string of 8 bits is usually transmitted at a time, which is equal to 1 byte. In contrast, in serial transmissions, only 1-bit is usually transmitted at a given time. Even though more bits are transmitted through parallel transmission, serial transmission is the most widely used in storage technologies. This is because disk controllers tend to handle only 1-bit at a given time, meaning that the disk's read-and-write head reads and writes 1-bit within the given time. The serial method of transmission has proved to be more pragmatic by eliminating overhead processing, signal skewing, and crosstalk. Today's most used serial interfaces in storage technologies, such as SATA, SAS, FC, and USB, utilize serial buses. Understanding storage protocols Storage protocols allow us to store and retrieve data in/from storage systems. The most used storage protocols are as follows: SCSI is a storage protocol that is used heavily in block-level storage systems. The server operating system uses the SCSI protocol to read and write data on a SCSI controller that manages storage devices. iSCSI places the standard SCSI protocol in an IP packet, thus extending its functionalities throughout the organization's network. FC is another way of extending the functionalities of the standard SCSI protocol, enabling storage consolidations and longer distances. FCoE does the same for the FC protocol as iSCSI does for the SCSI protocol. That being said, FCoE extends the functionalities of the FC protocol across Ethernet networks. Understanding file sharing protocols File sharing protocols enable data sharing over LANs, WANs, and the internet. The most used file sharing protocols are as follows: Server Message Block (SMB), also known as the Common Internet File System (CIFS), is a file sharing protocol used mostly by Windows OSes. Network File System (NFS) is a file sharing protocol used mostly by Unix and Linux. File Transfer Protocol (FTP) enables file sharing by transferring files from site to site. Hypertext Transfer Protocol (HTTP) enables file sharing over a World Wide Web (WWW) service. Secure Shell (SSH) enables remote file sharing over a secure connection. Understanding FC, HBA, and FC switches The HBA is an interface standard, whereas the FC switch is a network switch. Both components are compatible with FC, a high-speed network technology that is used to connect the two, thus creating the FC fabric. FC fabric consists of one or more FC switches, and, as such, it constitutes the SAN topology. What is iSCSI hardware? iSCSI, a block-level storage, uses an IP to send the SCSI commands over TCP/IP networks. iSCSI works in such a way that the clients, known as initiators, use the IP protocol to send SCSI commands called command descriptor blocks (CDBs) to storage devices known as targets. In SANs, the logical unit number (LUN) represents a logical disk. In iSCSI, TCP port 860 is reserved for the iSCSI system port, whereas TCP port 3260 represents iSCSI's default port. What is S2D? S2D, which was introduced in Windows Server 2016, is a feature that is supported in Windows Server 2019 too. It enables the grouping of disks into storage pools, thus creating software- defined storage called storage spaces. What is dedup? The idea behind the concept of dedup is to provide disk space savings. dedup is a technique that removes duplicated data from a dataset, thus storing a single copy of identical data on a disk. First, it analyzes the data to identify duplicated data in the dataset, and then the original file is stored in storage media while the duplicated files are replaced with a reference that points to the original file. What is storage tiering? storage tiering is a built-in feature in Microsoft's Windows Server, which enables the automatic transfer of the most frequently accessed files to faster storage. In simpler terms, this means it lets you combine high- performance storage with lowperformance storage (for example, the HDD and the SSD) to reduce storage costs. The storage tiering agents will place the most accessed files on the faster storage, while rarely accessed files are placed in slower storage. Managing storage with Server Manager and Windows PowerShell Prior to managing storage with Server Manager (servermanager.exe), ensure that the File and Storage Services role is added to the server. The following figure shows storage management using Server Manager on the local server. Managing storage with Server Manager and Windows PowerShell Other than Server Manager, you can manage storage with Windows PowerShell too. The following figure shows Windows PowerShell in action (powershell.exe) Understanding RAID Regardless of whether you have come across the term redundant array of independent disks or redundant array of inexpensive disks, you should know that you are dealing with the concept of fault tolerance. RAID is a technology that combines a considerable number of physical disks into a single logical unit so that it can protect data in the case of disk failure. RAID is not a backup solution and should never be considered as such. Types of RAID There are a considerable number of RAID types. The most used are as follows: RAID 0 is known as disk striping and offers higher read and write performance but it is not fault-tolerant. On Windows Server 2019, you can create a striped volume if you convert the disk from basic to dynamic. RAID 1 is known as disk mirroring and requires at least two disks for its implementation while offering excellent read and write performance. It works in such a way that all the data that resides on disk A is mirrored on disk B. In the case of disk failure, the RAID controller uses any of the available disks. RAID 5 is known as disk striping with parity and requires three disks for its implementation at a minimum. It represents the most fault-tolerant RAID that's available. The parity data is spread across all disks, meaning that RAID 5 can withstand the failure of a single disk. RAID 10 is known as the stripe of mirrors and combines RAID 1 with RAID 0, thus offering disk mirroring and striping at the same time. It requires four disks for its implementation at a minimum. In the case of disk failure, the rebuild time is very fast since the striping is spread across all drives. Hardware versus software RAID When it comes to deploying RAID, there are two types of RAID: Hardware RAID is an expensive solution and requires configuration prior to installing the OS. It is an electronic board that either the manufacturer of the server or you will plug into an adequate slot on the server's motherboard. Software RAID is a cheaper solution and is configured after the installation of an OS. It is an application that you will end up buying from a specific vendor. Understanding SDS If an organization, due to budget constraints, cannot afford to own NAS or SAN storage systems, then it can select the cheapest alternative that has to do with SDS. With Windows Server 2019, through S2D organizations can create virtualized networks with local storage. By doing this, they can build an SDS that helps separate the software that manages the storage from the storage hardware. Understanding resiliency using S2D The fault tolerance approach in S2D is called resiliency. It offers mirroring with parity, and so in terms of implementation, it is similar to RAID software. In Windows Server 2019, S2D offers fault tolerance and storage efficiency. Understanding high availability (HA) HA is a characteristic of a system that never fails, thus being available at all times. However, that is only possible in an ideal world. In our world, the highest available standard is 99.9999%. To achieve such a standard, we need to have numerous parameters in place. Thus, from backup to fault tolerance and from resilience to reliability, all storage media is required to be operational so that the system as a whole is highly available. Understanding high availability (HA) The following table shows the HA standards: Understanding disk types One reason why you should learn about disk types is so that you can get acquainted with their types and technical specifications. In turn, that will help you understand the real storage potential of any disk type. Understanding HDD The HDD is considered to be a secondary storage type, right after server memory (that is, RAM memory). As such, it is a computer component that uses the electromotor to spin the disk, and it contains a magnetic read-and-write head and metal platters that permanently store data. Each platter contains tracks and sectors. The starting point for storing data in HDDs is the outer track. The read-and-write head is located above the platter at a distance of microns, thus never touching the disk. Understanding HDD The data storage capacity is measured in bytes (nowadays, it is GB and TB), while the disc spinning speed is measured in rotations per minute (RPM). The most common RPM rates for PCs and laptops are from 5,400 RPM to 7,200 RPM, while for servers the most common RPM rates are from 10,000 RPM to 15,000 RPM. Usually, the HDD is located inside the computer's case and is mounted in drive bays. However, there are also external HDDs, which are mainly used for storing and backing up data. Understanding SSD The SSD is another storage technology that is considered to be secondary storage. Unlike HDDs, SSDs are memory chips with no moving parts. They use less voltage (usually 5V) than HDDs (12V for spinning the platters), are noiseless, more physically reliable, and provide faster data access. SSDs are behind HDDs in terms of capacity. However, based on current development trends in storage technologies, things might turn in favor of SSDs over HDDs. These days, many manufacturers offer their PCs and laptops with SSD disks that contain the OS and applications, as well as HDD disks to be used for data storage. Additionally, SSD drives are also encountered on servers, including NAS and SAN devices too. Understanding optical disk drives Unlike HDDs, which use an electromagnetic field to read and write data to/from disk platters, optical disk drives (ODDs) utilize laser beams with a specific wavelength to read and write data to/from compact discs (CDs). Always try to differentiate between the ODDs and optical discs (ODs), such as CDs or DVDs. The former is the device where the CD or DVD is inserted. CDs contain tracks in the form of a spiral. The starting point for storing data on CDs is the inner track. As is the case with HDDs, optical discs are measured by capacity in bytes. Usually, the capacity of common CDs is between 650 MB and 700 MB, while common DVDs range from 4.7 GB to 8.5 GB. In contrast to HDDs, the speed of optical discs is measured in KB/s and is determined by an x symbol that is equal to 150 KB/s. Therefore, if your optical drive has a speed of 24x, then its speed is 24 x 150 KB/s = 3600 KB/s = 3.6 MB/s. Understanding optical disk drives There are three recording types of optical disks, as follows: CD-ROM and DVD-RAM are read-only optical discs. CD-R and DVD-R/DVD+R are write-once optical discs. CD-RW and DVD-RW/DVD+RW are rewritable optical discs. The most common forms of optical discs at the time of writing are DVDs and Blu-ray discs. The latter has been designed to supersede DVD technologies, thus achieving tremendous capacities where a single layer holds 25 GB, a dual layer holds 50 GB, a triple layer holds 100 GB, and a quadruple layer holds 128 GB. Understanding basic disks Once you have installed the operating system on the server's hard disk, the hard disk structure is in its basic configuration. This means that the basic disk configuration is organized into partitions. The basic disk is based on the MBR and GPT partition schemes, and, as such, one partition cannot be extended on one or more physical disks. Instead, a partition can be extended by adding unallocated space from the same physical disk. Understanding dynamic disks To overcome the limitations of the basic disk, to be able to increase read-and-write performance with disk striping, and to operate with volumes instead of partitions, dynamic disk configuration was introduced. This means that volumes in dynamic disk configuration can be extended to more than one physical disk, thus allowing us to create five types of volumes: simple, mirrored, striped, spanned, and RAID-5 volumes. Understanding mount points When you attach an unallocated partition to a blank folder, what you have actually done is create a mount point. This allows us to increase the size of the folder if the partition where the folder is located is running out of space. In Windows Server 2019, you can use Disk Management (diskmgmt.msc) to create a mount point. Understanding filesystems The wellknown filesystems that are used by the Windows OS: File Allocation Table (FAT) is the earliest filesystem that was used by both MSDOS and Windows. As the name implies, it is based on a table that contains a map of clusters. A cluster is a unit of logical storage on the hard disk. FAT32 is the latest version of FAT. New Technology File System (NTFS) was introduced in the 1990s with Windows NT 3.1 and is still in use. Among the features that NTFS offers are disk quotas, Encrypting File System (EFS), journaling, and the Volume Shadow Copy Service (VSS). NTFS is a native filesystem on Windows Server 2019. Resilient File System (ReFS) was introduced in Windows Server 2012 and is supposed to be the successor of the NTFS. The new features that ReFS offers include resiliency, performance, and scalability. ReFS is available as a disk format option in Windows Server 2019. Extended File Allocation Table (exFAT) is a new version of FAT that was developed to be used primarily with USB flash drives and SD cards. It is interesting that exFAT is platform independent, thus enabling drives that have been formatted with this filesystem to be supported by Mac computers too. Understanding dynamic disks In Windows Server 2019, to convert a basic disk into a dynamic disk, complete the following steps: 1. Right-click the Start button. 2. Select Disk Management. 3. Right-click the preferred disk, and, from the context menu, select Convert to Dynamic Disk.... 4. If you have more than one disk, then, from the Convert to Dynamic Disk window, select the disks and click OK. 5. Click Convert in the Disks to Convert window. 6. After you have read the information in the Disk Management dialog box, click Yes. 7. Shortly after, the conversion will be completed. Mounting a VHD Based on the concept of the mount point, a mounted VHD drive is a mount point which, in this case, represents a drive that is mapped to an empty folder on a volume that uses the NTFS filesystem. Mostly, mounted VHD drives function like any other drives, except that instead of drive letters, they use the drive path. In Windows Server 2019, you can attach a VHD to your server using Disk Management. To do so, complete the following steps: 1. Press the Windows key + R. 2. In the Run window, enter diskmgmt.msc and then hit Enter. 3. In the Disk Management window, click the Action menu and select Attach VHD. 4. Shortly after, the attached VHD will be displayed in the Disk Management window. Understanding Distributed File System (DFS) If you have ever wondered how you can share data from your servers in an authorized and controlled way, then look no further than DFS. With DFS, data that is stored in shared folders that's located on different servers can be grouped into logically structured namespaces. This makes it possible for users to access the data as if it were stored on local computers. Understanding Distributed File System (DFS) In Windows Server 2019, DFS is part of the File and Storage Services role. Thus, to install DFS on your server, you should expand the File and Storage Services role, expand File and iSCSI Services, and then select DFS Namespace, DFS Replication, and File Server Resource Manager. Exercise: enabling dedup on Windows Server 2019 Refer to the textbook from page 288 to 290.