Sorry, the language specified is not available for this page

    // Blog

    AWS Elastic Block Storage GP3: Fewer Trips Around The World at a Lower Cost

    May 4, 2021 | By Sungard AS |

    Four years ago, IDC issued a report on the soaring growth of the world’s total amount of data, estimating that by 2025 the collective sum would skyrocket from 33 zettabytes (ZB) in 2017 to 175 ZB in 2025. (NOTE: 1 zettabyte = 1 trillion gigabytes.) In other words, if you were to store 175 ZB on a typical DVD, your stack of discs would encircle the planet 222 times. If you wanted to download all that data at the average current internet connection speed, it would take around 1.8 billion years to download the entire collection. That’s a LOT of data. 

    Fortunately, there are several ways to store all that data in efficient methods. The storage “ecosphere” falls into three layers

    1. Core: which includes data centers and cloud data centers
    2. Edge: which encompasses things like branch offices and cell towers
    3. Endpoints: where devices we use every day come into play – smartphones, PCs, and Internet of Things (IoT) devices

    Businesses typically store their data in the first layer, the core – where hundreds of data centers and millions of equipment racks house data in blocks or fixed-size raw storage units, much like fixed disk drives.  

    Amazon Web Services (AWS) offers one type of storage solution – Amazon Elastic Block Store (EBS) – a useful block storage service that works in conjunction with the Amazon Elastic Computer Cloud (EC2) for intensive workloads. Introduced in 2008, Amazon EBS takes the name “elastic” from the fact that the block store can adapt volume size to an application's current needs, allowing the user to pay only for what they need.  

    In 2014, Amazon introduced gp2 to offer general-purpose solid-state drive (SSD) performance at a very low price. It was a simple and cost-effective way to meet the growing performance and throughput requirements of many applications including virtual desktops, medium-sized databases such as SQLServer and OracleDB, and dev/test environments. 

    Amazon EBS gp3 is the newest generation of Amazon EBS that enables customers to provision performance independent of storage capacity, providing a 20 percent lower price point per GB than existing gp2 volumes. With gp3 volumes, customers can scale IOPS (input/output operations per second) and throughput without needing to provision additional block storage capacity – again, paying only for the resources they need and use.  

    The savings start to add up 

    Breaking down the price savings, if we compare the pricing between gp2 and gp3 we can see the where the 20 percent savings comes from:  

    • General Purpose SSD (gp2) Volumes = $0.10 per GB-month of provisioned storage  
    • General Purpose SSD (gp3) Storage = $0.08/GB-month  

    On the surface, it’s a no-brainer to switch to gp3 – but the savings don’t end there. Breaking down the performance savings, EBS gp2 has been the go-to volume type for most EC2 instances for many years. With its baseline performance of three IOPS per GiB, a 100 GiB volume would only get a guaranteed 300 IOPS, with some capacity to burst to 3,000. The cost of that volume would be $10/month. However, to guarantee 3,000 IOPS all the time we had only two choices:  

    1. Use a provisioned IOPS volume type (io1): The cost of running a 100 GiB volume with 3,000 IOPS would be $208/month (US-East-1); OR  
    2. Overprovision capacity: If we provision a 1,000 GiB gp2 volume then we would also get a guaranteed baseline of 3,000 IOPS at the cost of $100/month.  

    While the obvious answer is #2, it feels wasteful to have so much unused storage. Plus, if your volumes require backups, the additional cost of storage for snapshotting a terabyte volume vs. 100 GiB eats away the 50 percent cost savings that option #2 made over option #1. With gp3 we have another option: use a gp3 volume type. The cost of running a 100 GiB volume with 3,000 IOPS would be $8/month (US-East-1).  

    The performance advantage with gp3 volumes is the ability to increase the IOPS and throughput separate from the EBS volume size. If you need IOPS, then gp3 is the way to go as you have a baseline performance of 3,000 IOPS and can scale to 16,000 IOPS, regardless of volume size. 

    Put this way, you can see how much you would pay for a 100GiB volume with 3,000 IOPS of guaranteed performance: not $208/month, not $100/month, but $8/month.  

    How to switch from gp2 to gp3

    If you are currently using gp2, you can easily migrate your EBS volumes to gp3 using Amazon EBS Elastic Volumes, an existing feature of Amazon EBS. Elastic Volumes allows you to modify the volume type, IOPS, and throughput of your existing EBS volumes without interrupting your Amazon EC2 instances. In fact, you don’t even need to power down your instance; just change the volume type in the AWS Management Console and it will be seamlessly migrated without any downtime.   

    When you create a new Amazon EBS volume, Amazon EC2 instance, or Amazon Machine Image (AMI), you can now choose the gp3 volume type.  

    Sungard Availability Services (Sungard AS), a recognized leader with a 40-year track record providing customers with highly available hybrid solutions, recommends that all customers consuming gp2 storage begin to investigate the benefits of gp3. Fortunately, Sungard AS’ NextGen Managed Public Cloud team is already working with customers today to understand whether the move to gp3 is right for their workloads. 

    If you have any questions, we're ready to help.

    Get in touch with a Sungard AS sales specialist.

    Share Post: