Databases form the foundation of many business systems and tend to be very widespread across the enterprise. In fact, it's quite common to have several databases with varying workloads running on multiple physical servers. Too many database instances on too many machines leads to issues such as server sprawl, high power consumption, increased data center footprint, inefficient resource usage and complex management.
Consolidating several applications on a single, large server with strong computing power allows you to leverage the advantages of virtualization and improve cost efficiency through higher utilization of resources, while addressing the space, energy and manageability issues mentioned above.
Drivers for Consolidation
- Reducing cost – Through consolidation, servers will run closer to capacity, reducing inefficiencies and allowing for fewer larger machines. Reducing capital and operating expenditure is one of the biggest factors driving companies to consolidate. Upgrading to fewer machines and newer hardware allows for reductions in rack space, power and cooling needs. Database sprawl is also minimized because the machines are more easily centrally managed. Central management provides better control, reducing administrative and maintenance overhead, and it can also help reduce licensing costs. For more information, read the Business Value of 4- and 8-Socket Scale-Up Deployments blog.
- Application performance – Some application and workloads lend themselves particularly well to scale-up rather than scale-out deployments. Factors contributing to a decision to consolidate rather than scale out include processing and memory requirements, the number of end users, scalability cost and administrative cost. Consolidating to larger servers allows you to add CPU, memory, I/O and storage to your servers, and it does not require you to re-write your application to harness the new horsepower. For large databases, scale-up architectures can provide higher levels of scalability than large numbers of scale-out distributed databases, and scale-up servers are often easier and less expensive to manage. For more information, read the Applications Drive 4- & 8-Socket Scale-Up Deployments blog.
- Standardization and centralization – One of the issues in database sprawl is an inconsistent approach to database structure and database management. Consolidation can be used to pull these various databases into a centrally managed system. By having a common set of requirements and methodologies, administrators are able to take advantage of predictable workflows for patching and configuration, better audit control over security configuration and streamlined hardware requirements. This also increases the opportunity to make improvements in ease of deployment and provisioning as well as application interoperability.
- Lack of space – Lack of space in the data center is one of the more direct reasons for needing to consolidate. As companies grow, so too do their hardware needs, and with them, the need for someplace to store all of that hardware. Acquiring new space through expansion of the data center or the creation of new data centers can result in significant capital expenditure. Clients should focus on enabling consolidation by upgrading to newer larger servers to take advantage of higher performance and scalable growth.
- Agility – Consolidation helps with building a long-term dynamic and power-aware IT infrastructure that allows for better control and flexibility of computing resources in terms of their placement, sizing and overall utilization. If you move applications onto newer hardware, those applications can take advantage of improved performance and reliability from the new machines, which can be better configured for high-availability scenarios, reducing downtime and allowing for rolling upgrades.
- Green IT – Focusing on providing a “greener” operating environment is similar to the motivations for reducing cost and increasing efficiency, but the ultimate goal is environmental benefit rather than cost savings. Consolidation plays an important role here in reducing the data center footprint. Fewer computers and fewer idle machines result in lower power consumption and a reduced need for cooling. New hardware can also provide better energy efficiency as well, because it can take advantage of more power-efficient technologies.
Why Lenovo X6 is the Perfect Consolidation Server
- High computing power – The Lenovo x3850 X6 and x3950 X6 have a tremendous amount of computing power. Database engines can leverage available cores for parallel query processing achieving faster query execution time. The higher processor cache improves query processing while reducing the latencies involved in fetching the data from the main memory. When fully loaded, the Lenovo x3950 X5 provides up to 144 cores, which enables customers to achieve better consolidation factor in migrating and consolidating many databases on to a single host, thereby achieving reduced data center footprint and increased power and cooling savings. For more information, read the Lenovo X6 Server Consolidation – Save 80% by Consolidating HP Servers blog and the Lenovo x3950 X6 Achieves World-Record SPECvirt Benchmark blog.
- Scalability and agility – Lenovo X6 servers provides seamless scalability to meet increasing workload demands and allow clients to scale from two processors all the way up to eight processors. The agility and adaptability of the Lenovo X6 modular rack design enables you to design a fit-for-purpose solution that meets your needs. You can also realize infrastructure cost savings by hosting multiple generations of technology in a single platform.
- Unmatched reliability and availability – To ensure business continuity and increase end-user productivity, it is imperative that you maximize uptime of server hardware and server operating systems. All Lenovo servers have strong reliability, availability and serviceability (RAS) capabilities, but the Lenovo System x3850 X6 and x3950 servers have advanced RAS features not found in other servers. The differentiated X6 self-healing technology proactively identifies potential failures and transparently takes necessary corrective actions. The Lenovo X6 servers provide mainframe-like RAS because they integrate across the hardware and software stack. For more information, read The Only Good Downtime is NO Downtime blog and the Supporting Mission-Critical Workloads blog.
- High-speed and high-capacity memory – The capacity and speed of the server’s main memory are crucial to memory-intensive applications such as databases. The Lenovo x3850 X6 supports up to 6 TB of memory and the x3950 X6 support up to 12 TB of memory. This massive memory capacity enables the X6 servers to support more database virtual machines seamlessly on a single host, thereby achieving a better database consolidation factor. For more information, read the Optimizing Memory Performance blog.
- Most PCIe slots for scalable networking and storage access – I/O-intensive workloads running on the physical host can easily saturate the underlying networking cards. Therefore, it is essential that the host support enough PCIe slots. This enables customers to start with a minimum configuration and expand network bandwidth as required. The x3950 X6 supports an unmatched 22 PCIe slots plus two dedicated mezzanine LOM slots, which provides you the flexibility to get started with a minimal number of network cards and then scale up as more virtual machines are consolidated on the server. This helps to reduce network latencies and improve network bandwidth.
- SSD I/O performance – X6 servers provide multiple disk backplane options to give you greater flexibility. The x3950 X6 supports up to 16 SAS solid-state drives (SSDs) (x3850 – eight) providing reliable and higher-performing storage than traditional hard-disk drives (HDDs). The x3950 X6 supports up to eight (x3850 X6 – four) NVMe Enterprise Performance SSDs. These hot- pluggable 2.5-inch front-loading SSDs provide more IOPS over traditional SSDs with lower latencies. This can be used to store the I/O-intensive server database components. For more information, read the Advantages of Solid-State Drives blog.
For more information on each system, check out the web sites: Lenovo System x3850 X6, a 4U rack server scalable to four processors and the Lenovo System x3950 X6, an 8U rack server scalable to eight processors.