MarkLogic Data Hub Service Pricing
Low and Predictable Cost for Variable Workloads
As a cloud-neutral, serverless data hub, Data Hub Service offers consumption-based pricing based on cloud credits. You can pay on the go or get savings by buying credits in bulk in advance. The pricing is all-inclusive (includes hardware, software, operations, provisioning, and 24/7 support) and simple – you pay for compute, storage, and bandwidth consumed.
Workload-Aware Scaling
Independently scales curation, operational, and analytical workloads for high performance and reliability.
Free Bursting
Automated, free compute burst lets you meet SLAs for workload with variable demand without cost overrun.
Predictable Cost
Plan for average demand and utilize accumulated, unused, compute credits for unpredictable workloads.
How Do You Consume and Accumulate MarkLogic Capacity Units (MCU)?
Predictable Pricing
Data Hub Service offers simple consumption-based pricing. There is no infrastructure to manage and you pay a fixed rate for compute (billed as MarkLogic Capacity Units, or MCUs) and the storage consumed.
Data Hub Service is available on AWS and Azure. There are three ways to pay for Data Hub Service instances: Low Priority Fixed, Standard On-demand, and Standard Reserved.
Low Priority Fixed
Fixed compute capacity for intermittent workloads
| Compute | Storage |
|---|---|
| $0.15 | $0.10 |
| /MCU per hour* | /GB per month* |
Details
- Fixed compute capacity by the hour
- Does not provide high-availability and auto-scaling
Recommended for
- Batch (or ad-hoc) workloads
- Development or test instances
- Workloads with flexible start and end times
Example
A low-priority fixed instance for 8 MCUs, provisioned in Azure US East region, that runs 5 hours for 20 days with a storage footprint of 32 GB will cost $123/month (excluding data transfer fee).
Standard On-demand
On-demand compute capacity for 24×7 availability needs
| Compute | Storage |
|---|---|
| $0.25 | $0.10 |
| /MCU per hour* | /GB per month* |
Details
- Elastic compute capacity by the hour
- Offers high-availability and free compute bursting (or auto-scaling)
- Stores up to 12x of unused compute credits that are used during burst for free
Recommended for
- Workloads that require continuous availability
- Unpredictable workloads
- Strict service level agreements
Example
A standard on-demand instance for 32 MCUs, provisioned in AWS US East region, that runs 24 hours for 30 days with a storage footprint of 128 GB will cost $5773/month (excluding data transfer fee).
Standard Reserved
On-demand compute capacity for 24×7 availability needs
| Contact Sales |
Details
- Elastic compute capacity by the hour
- Offers high-availability and free compute bursting (or auto-scaling)
- Stores up to 12x of unused compute credits that are used during burst for free
Recommended for
- Workloads that require continuous availability
- Unpredictable workloads
- Strict service level agreements
Example
A standard on-demand instance for 32 MCUs, provisioned in AWS US East region, that runs 24 hours for 30 days with a storage footprint of 128 GB will cost $5773/month (excluding data transfer fee).
Faster and Better than Doing It Yourself
Data Hub Service (DHS) customers benefit from accelerated time to value and faster innovation cycles. With Data Hub Service, MarkLogic owns the Ops and you simply pay for compute, storage, and bandwidth consumed by your workloads. This is in contrast to you running the MarkLogic Data Hub in a public or private cloud, where you own the Ops and have complex, multi-line billing for licensing, infrastructure, and administration costs.
Top FAQs
Is there a performance increase with increased capacity?
As an elastic, serverless cloud data hub, Data Hub Services scales linearly with provisioned compute capacity (MCU). The higher baseline capacity not only results in higher performance (i.e. transactions per second) but also enables auto-handling of peak load demand for a longer time duration at no extra cost due to free compute bursting. This makes it easy for customers to easily optimize the baseline compute capacity based on ongoing load profile and the desired SLAs.
For detailed testing results on Data Hub Service scalability, please check out the following video.
How do I determine how many MCUs to buy?
There is no magic formula to sizing the baseline compute capacity of Data Hub Service. It depends on multiple factors: data volume in the data hub, data ingestion velocity, query load, query complexity, etc. The Data Hub Service pricing sheet has some heuristics to estimate baseline compute capacity based on data volume but it does not account for every factor.
To size the capacity of Data Hub Service for your particular use-case, we recommend that you work directly with your MarkLogic sales team. They can help you find the equilibrium that maximizes free compute bursting while not having to pay and provision capacity to handle peak demand.
Sign Up for Our Live Demo
See how MarkLogic integrates data faster, reduces costs, and enables secure data sharing.