Enterprise
On-Premises and Hybrid Cloud Data Warehousing
Yellowbrick Data Warehouse offers a hybrid cloud data warehousing solution combining the performance and density advantages of on-premises and the cost and unlimited scale of cloud-based storage solutions. With Yellowbrick, you can store and manage your data in a flexible, scalable, and cost-effective manner, while also benefiting from advanced analytics capabilities. You can choose to run Yellowbrick in either public cloud, on-premises or both.
Andromeda: Optimal Performance for On Premises Data Warehousing
For on-premises use cases, Yellowbrick has developed the “Andromeda” server hardware instance, driving new efficiencies in price-performance.
Optimized for Yellowbrick’s database, Andromeda is designed to bring significant performance, efficiency, and economic advantages to customers deploying Yellowbrick inside private clouds.
Unmatched Query Throughput and Cost Efficiency
With Yellowbrick’s database and Andromeda, it’s not uncommon to find one server node providing the equivalent query throughput of a dozen or more nodes of competitive cloud and on-premises databases, at a fraction of the total cost.
Andromeda-optimized hardware instances are designed to bring significant performance, efficiency, and economic advantages to customers deploying Yellowbrick inside private clouds.
The Best of On-premises and Cloud Environments
The result is a new kind of cloud-compatible data warehouse that provides the best economics in the industry, along with all other expected features and functions of a mature product that can be trusted to help run your business faster and more efficiently.
For more details about Andromeda, see our Andromeda Optimized Instances whitepaper.
Instance Design for Data Warehousing
Andromeda is a blade server based on an existing design customized for Yellowbrick by a large server original design manufacturer (ODM) that supplies several public cloud vendors. The server motherboards are manufactured and tested on the same assembly lines that produce servers for many major original equipment manufacturers (OEMs) and ODMs.
COMPUTE
For compute, we care about the cost of each CPU core, which largely dictates how fast we can go on executing instructions, and the cost per memory channel, which largely dictates how fast we can do large aggregates, joins, and sorts.
The introduction of AMD’s EPYC processors makes it affordable to acquire 64 cores of compute with eight memory channels, resulting in the lowest possible price per core and memory channel.
NETWORK
100Gb networks are now the sweet spot in cost per unit of bandwidth. Since a redundant network architecture is required for high availability, each server node has access to two network interfaces running over two separate switches.
Yellowbrick makes use of the features on the EPYC processor and the network interface to closely couple the fabric and query processing, enabling us to drive an incredible 200Gb/sec per node of data across the network – roughly 20GB/sec per node, full duplex, or 400GB/sec per chassis.
To make this process efficient, we use a remote direct memory access (RDMA) fabric that allows direct movement of data – typically cache-resident – between nodes, with no TCP/IP or Linux kernel in the way to slow things down.
STORAGE
Each Andromeda server supports 8x 7mm NVMe U.2 drives, offering 24GB/sec of read bandwidth per node and 16GB/sec of write bandwidth. Because data is compressed, the effective read bandwidth per node is over 3x higher, sometimes peaking at over 100GB/sec of user data scanned per server node.
RESILIENCE
Within the Andromeda chassis, all the following components are both hot-swappable and redundant:
•SSDs
•Whole server blades
•Network switches
•Power supplies
•Fans
Andromeda has been tested to scale efficiently from 3 blades to 80 blades (8x chassis) per data warehouse instance. The tables below list key Andromeda specifications and configurations:
CATEGORY | SPECIFICATION |
|---|---|
CPU | AMD EPYC with 64 cores, 8 memory channels |
Network | 200Gb (2x100Gb) RDMA fabric and 2x switches |
Storage | 8x U.2 hot-swap NVMe SSDs |
Storage Capacity Per Blade | 16TB, 32TB, 64TB |
Memory Per Blade | 512GB, 1TB |
Minimum Blades | 3 (in 1 chassis) |
Maximum Blades | 80 (8 chassis; 10 blades each) |
Minimum vCPUs | 384 (3 blades, 1 chassis) |
Maximum vCPUs | 10,240 vCPU (80 blades, 8 chassis) |
SINGLE CHASSIS 8U | 2-CHASSIS 14U | 3-CHASSIS 20U | 4-CHASSIS 26U | 8-CHASSIS 50U | |
|---|---|---|---|---|---|
Compute Nodes | 3, 4, 6, 10 | 20 | 30 | 40 | 80 |
vCPU | 384, 512, 768, 1280 | 2560 | 3840 | 5120 | 10240 |
User Data (TB) VD Models | 45, 90, 180, 375 | 750 | 1125 | 1500 | 3005 |
User Data (TB) ED Models | 90, 185, 375, 750 | 1500 | 2255 | 3005 | 6015 |
User Data (TB) FD Models | 185, 375, 750, 1500 | 3005 | 4510 | 6015 | 12030 |
Memory (TB)* | 1.5, 2, 3, 5.1 | 10.2 | 15.3 | 20.4 | 40.8 |
*2x expanded memory option available. Minimum node count = 3; maximum node count = 80. User data assumes 3.6X data compression.
SINGLE CHASSIS 8U | 2-CHASSIS 14U | |
|---|---|---|
Compute nodes | 4 | 8 |
Model | 033-FE-104 | 033-FE-208 |
CPU Type | AMD 7700 64-core | AMD 7700 64-core |
vCPU | 512 | 1,024 |
Raw Space (TB) | 245.8 | 491.5 |
Memory (TB) | 4 | 15.3 |
Networking | 200Gb (2x100Gb) RDMA fabric and 2x switches | 200Gb (2x100Gb) RDMA fabric and 2x switches |
Drives per Node | 8x U.2 hot-swap NVMe SSDs | 8x U.2 hot-swap NVMe SSDs |
Manager Nodes | 2x (fully redundant, HA) | 2x (fully redundant, HA) |
Power – Peak (Watts) | 2,700 | 4,700 |
Thermal – Peak (BTU/hr) | 9,213 | 16,037 |
Weight (kgs) | 94 | 118 |
Rackspace Dimensions (HxWxD) | 14” x 17.6” x 31.25” | 24.5″ x 17.6″ x 31.25″ |
Operating Temperature | 50°F – 95°F (10°C – 35°C) | 50°F – 95°F (10°C – 35°C) |
Power Requirements | 208VAC-240VAC @ 8.3A – 60A | 208VAC-240VAC @ 8.3A – 60A |
Safety | UL 60950-1, CAN/CSA-C22.2 No. 60950-1,EN 60950-1, IEC 60950-1 | UL 60950-1, CAN/CSA-C22.2 No. 60950-1,EN 60950-1, IEC 60950-1 |
Emissions | FCC Part 15 Class A, CISPR 22/CISPR 24 Class A, EN55032/55024 Class A | FCC Part 15 Class A, CISPR 22/CISPR 24 Class A, EN55032/55024 Class A |
Encryption | Data-at-rest encryption included | Data-at-rest encryption included |
Full-Service Support
The world’s only data warehouse for hybrid and multi-cloud environments gives healthcare providers, pharma, and biotech companies the price/performance, agility, and flexibility they need to improve care and financial outcomes in the face of massive data challenges.
Yellowbrick is the world’s fastest data warehouse for hybrid and multi-cloud environments enhancing global supply chain management with better, faster analytics, including real-time speed, petabyte-scale deep analytics, and industry-leading deployment flexibility
With Yellowbrick, media, advertising, and entertainment providers the price/performance, agility, and flexibility they need to improve and deliver memorable user experiences in the face of increasing competition for user attention. Capture more first-party data to understand customer behavior better and deliver more targeted and relevant content.
Government agencies are working under controlled budget constraints. Data is continuing to grow at exponential rates. More complex analytics are needed on this fast-expanding data at any location. Yellowbrick is the solution.
Resources
