Quick Reference
A concise reference for all 14 workload types in Lakemeter. Use this page to quickly identify which workload type to use and what configuration options are available.
Workload types at a glance
| Workload | What it is | When to use it | Min. Tier |
|---|---|---|---|
| Jobs | Batch compute for ETL, ML training, data processing | Scheduled or triggered pipelines that run and terminate | Standard |
| All-Purpose | Interactive compute for notebooks and development | Ad-hoc analysis, prototyping, development clusters | Standard |
| DLT | Delta Live Tables declarative data pipelines | Managed ETL with built-in data quality and monitoring | Standard |
| DBSQL | SQL analytics warehouses | BI dashboards, SQL queries, analyst workloads | Standard (Classic/Pro), Premium (Serverless) |
| Model Serving | Real-time ML model inference endpoints | Deploying custom ML models for online predictions | Premium |
| Vector Search | Managed vector database | Similarity search, RAG applications, embeddings | Premium |
| FMAPI (Databricks) | Foundation Model API for open-source models | Llama, DBRX, Mixtral via Databricks-hosted endpoints | Premium |
| FMAPI (Proprietary) | Foundation Model API for third-party models | GPT-4, Claude, Gemini via Databricks gateway | Premium |
| Lakebase | Managed PostgreSQL-compatible database | Transactional workloads, application backends | Premium |
| Databricks Apps | Managed app hosting on Databricks | Hosting web applications, dashboards, internal tools | Premium |
| AI Parse (Document AI) | AI-powered document parsing and extraction | Processing PDFs, invoices, contracts with AI | Premium |
| Shutterstock ImageAI | AI image generation via Shutterstock | Generating images for marketing, content, prototyping | Premium |
Key terms
| Term | Meaning |
|---|---|
| DBU | Databricks Unit -- the billing unit for Databricks services. Different workload types have different $/DBU rates. |
| SKU | Stock Keeping Unit -- identifies the specific product being priced (e.g., JOBS_COMPUTE, SERVERLESS_SQL_COMPUTE). |
| Photon | Hardware-accelerated query engine. Doubles the DBU rate but often halves runtime for compatible workloads. |
| Serverless | Databricks-managed infrastructure. No VM configuration needed; infrastructure cost is included in the DBU price. |
| Classic | Customer-managed infrastructure. You choose instance types and pay DBU + VM costs separately. |
Configuration by workload type
Compute workloads (Jobs, All-Purpose, DLT)
These workloads share a common configuration pattern:
| Field | Description | Default |
|---|---|---|
| Serverless | Toggle on for Databricks-managed compute, off for classic | Off |
| Serverless Mode | Standard or Performance (Performance uses 2x DBU rate) | Standard |
| Photon | Hardware-accelerated engine (classic mode only) | Off |
| Driver Instance Type | VM instance for the driver (classic only) | -- |
| Worker Instance Type | VM instance for workers (classic only) | -- |
| Worker Count | Number of worker nodes (classic only) | 2 |
| Driver Pricing Tier | On-Demand, 1-Year Reserved, 3-Year Reserved | On-Demand |
| Worker Pricing Tier | Spot Instances, On-Demand, 1-Year Reserved, 3-Year Reserved | Spot Instances |
| Payment Option | No upfront, partial upfront, all upfront (reserved only) | No upfront |
DLT-specific (classic mode only):
| Field | Description | Options |
|---|---|---|
| SDP Edition | Feature tier for the pipeline (Core, Pro, Advanced). Hidden when Serverless is enabled. | Core, Pro, Advanced |
Usage fields (Jobs):
| Field | Description | Default |
|---|---|---|
| Runs Per Day | Number of job executions daily | 1 |
| Avg Runtime (minutes) | Average duration of each run | 30 |
| Days Per Month | Active days per month | 22 |
Usage fields (All-Purpose, DLT):
| Field | Description | Default |
|---|---|---|
| Hours Per Month | Total compute hours | -- |
DBSQL
| Field | Description | Options |
|---|---|---|
| Serverless | Checkbox to enable Serverless SQL (Premium+ only). When off, choose Pro or Classic. | On (Serverless) |
| Size | Determines DBU/hr consumption | Small (12 DBU/hr) |
| Number of Clusters | Concurrent cluster count for scaling | 1+ |
| Hours Per Month | Warehouse uptime | -- |
DBSQL warehouse sizes and DBU rates:
| Size | DBU/hr |
|---|---|
| 2X-Small | 4 |
| X-Small | 6 |
| Small | 12 |
| Medium | 24 |
| Large | 40 |
| X-Large | 80 |
| 2X-Large | 144 |
| 3X-Large | 272 |
| 4X-Large | 528 |
Model Serving
| Field | Description | Options |
|---|---|---|
| Endpoint Type | Compute tier for the endpoint | CPU, GPU Small (T4), GPU Medium (A10G 1x), GPU Large (A10G 4x), etc. |
| Number of Endpoints | Concurrent endpoint instances | 1+ |
| Hours Per Month | Endpoint uptime | Default: 730 (24/7) |
Vector Search
| Field | Description | Options |
|---|---|---|
| Vector Search Type | Standard (4 DBU/hr per 2M vectors) or Storage Optimized (18.29 DBU/hr per 64M vectors) | Standard |
| Capacity (M vectors) | Number of vectors in millions | 1 |
| Storage (GB) | Additional storage capacity (20 GB free per endpoint unit, $0.023/GB/mo above) | 0 |
| Hours Per Month | Service uptime | Default: 730 (24/7) |
FMAPI (Databricks models)
| Field | Description |
|---|---|
| Model | Databricks-hosted model (e.g., Llama, DBRX, Mixtral) — LLMs and Embedding models |
| Rate Type | Token-based (input token, output token) or Provisioned (provisioned scaling, provisioned entry) |
| Quantity | For token-based: millions of tokens/month. For provisioned: hours/month (730 = 24/7) |
FMAPI (Proprietary models)
| Field | Description |
|---|---|
| Provider | OpenAI, Anthropic, or Google |
| Model | Specific model (GPT-4, Claude, Gemini, etc.) |
| Endpoint Type | Global or In-Geo |
| Context Length | All, Short, or Long (affects pricing for some models) |
| Rate Type | Input Token, Output Token, Cache Read, Cache Write, Batch Inference, Provisioned Scaling (availability depends on model) |
| Quantity (millions) | Token volume |
Lakebase
| Field | Description | Default |
|---|---|---|
| Capacity Units (CU) | Compute units: 1, 2, 4, or 8 CU | 1 |
| Number of Nodes | 1 = primary only, 2-3 = primary + read replicas (HA) | 1 |
| Storage (GB) | Database storage capacity (0-8192 GB) | 0 |
| Hours Per Month | Instance uptime | Default: 730 (24/7) |
Databricks Apps
| Field | Description | Default |
|---|---|---|
| App Size | Medium or Large | Medium |
| Hours Per Month | App uptime | Default: 730 (24/7) |
AI Parse (Document AI)
| Field | Description | Default |
|---|---|---|
| Mode | Pages-based or direct DBU | Pages |
| Complexity | Low (text), Low (images), Medium, High — affects DBU rate per 1000 pages | Medium |
| Pages (thousands) | Number of pages to process in thousands | -- |
Shutterstock ImageAI
| Field | Description | Default |
|---|---|---|
| Images Per Month | Number of images to generate | -- |
Cost formula summary
| Workload type | Cost formula |
|---|---|
| Classic compute (Jobs, All-Purpose, DLT) | (DBU/hr x hours/month x $/DBU) + (VM $/hr x hours/month) |
| Serverless compute (Jobs, All-Purpose, DLT, DBSQL) | DBU/hr x hours/month x $/DBU (VM cost included) |
| DBSQL Classic/Pro | (Warehouse DBU/hr x clusters x hours/month x $/DBU) + VM costs |
| Model Serving | Endpoint DBU/hr x endpoints x hours/month x $/DBU |
| Vector Search | Endpoint units x DBU/unit x hours/month x $/DBU |
| FMAPI | Token quantity (M) x DBU per M tokens x $/DBU |
| Lakebase | (CU x nodes x hours/month x $/DBU) + storage costs |
| Databricks Apps | App DBU/hr x hours/month x $/DBU |
| AI Parse | Pages (thousands) x DBU per 1000 pages x $/DBU |
| Shutterstock ImageAI | Images x fixed cost per image |