[Yandex Cloud documentation](../index.md) > [Yandex Managed Service for YTsaurus](index.md) > Pricing policy # Yandex Managed Service for YTsaurus pricing policy This section describes the Managed Service for YTsaurus pricing [policy](#rules) and [current prices](#prices) for its resources. For other Yandex Cloud resources consumed while using the service you pay as per their pricing plans. The prices for service products are also available in the [price list](https://yandex.cloud/en/price-list). {% note info %} Currency of Service rates (prices) depends on the company you made a contract with: * Prices in US dollars are applicable to customers of Iron Hive doo Beograd (Serbia) or Direct Cursus Technology L.L.C. (Dubai). * Prices in Russian roubles are applicable to customers of Yandex.Cloud LLC. {% endnote %} All prices in RUB and KZT are inclusive of VAT; in USD, net of VAT. ## Cluster status {#running-stopped} Pricing applies differently depending on the cluster status: * For a `Running` cluster, you pay for both the computing resources and storage size. * For a `Stopped` cluster, you only pay for your storage size. ## What goes into the cost of using Managed Service for YTsaurus {#rules} When using Managed Service for YTsaurus, you pay for: * Disk type and storage size. * Computing resources allocated to the cluster components. * Number of vCPUs. * Amount of RAM. * Outgoing traffic from Yandex Cloud to the internet. In all calculations: 1 GB = 230 bytes, 1 MB = 220 bytes. ### Using cluster components {#rules-components-uptime} The component operation cost is charged per hour based on what computing resources you allocate for it. You can check the supported component configuration options in the [relevant section](concepts/component-types.md) and see prices for using vCPUs and RAM in the [Prices](#prices) section. You can choose the configuration option for exec and tablet nodes based on the expected load. The number of service components, such as master, storage, system, and proxy nodes, depends on the cluster configuration and is applied automatically. See below for the minimum setup of service components required for a YTsaurus cluster to run: * **Master node**: Three configuration instances, 4 vCPUs and 12 GB RAM. * **System node**: Two configuration instances, 24 vCPUs and 64 GB RAM. * **Proxy node**: Two instances: One HTTP proxy and one RPC proxy. * Service **tablet node**: Two `c8-m16` configuration instances, 8 vCPUs and 16 GB RAM. Service tablet nodes are used for system dynamic tables. * **Storage node**: The number of storage nodes depends on the storage configuration and is applied as per below. **Defining the number of storage nodes in a cluster** 1. For each `HDD`, there is one storage node with 4 vCPUs and 8 GB RAM. 1. For each `SSD` or `non-replicated SSD`, there is one storage node with 8 vCPUs and 8 GB RAM. Additionally, there are three service storage nodes with 8 vCPUs and 8 GB RAM. They are used to store chunks and logs of system dynamic tables. The minimum billing unit is one minute (for example, 90 seconds of the component operation cost is same as two minutes). ### Disk space usage {#rules-storage} You pay for the following: * The size of the storage for the exec node, where various jobs are run. * The size of the storage for the service storage nodes that stores chunks and logs of system dynamic tables. This size is applied automatically, with three non-replicated SSDs, 93 GB each. * The size of storage for the storage node that houses the data of all YTsaurus objects that are created during its operation. This storage may consist of one or two disks. The number and size of such disks are defined as per below. **Defining the number of disks in storage nodes** 1. For each storage node, there is a main disk with the type and size you specify; this disk is used to store chunks. 1. For each storage node for an `SSD` or `non-replicated SSD`, there is a service disk of the same type; this disk is used to store the logs and snapshots of dynamic tables. 1. The service disk size depends on the storage configuration and is calculated as follows: * Assuming the storage node is for an `SSD` with `N` size, the service disk size is `0,1 × N` GB, rounded up to the nearest integer. * Assuming the storage node is for a `non-replicated SSD` with `N` size, the service disk size is `0,1 × N`, rounded up in increments of 93 GB. The minimum billing unit is 1 GB per minute; e.g., storing 1 GB for 1.5 minutes is billed as 2 minutes. ### Cluster cost calculation example {#example} Suppose we calculate the 30-day usage cost of a cluster with the following configuration: **Configurable components:** * **Exec node**: One `c8-m32` configuration instance: 8 vCPUs and 32 GB RAM. * **Tablet node**: Three `c8-m16` configuration instances, 8 vCPUs and 16 GB RAM. * **Storage** with the following configuration: * `HDD` storage: Three disks, 2048 GB each. * `Non-replicated SSD` storage: Three disks, 465 GB each. **Service components:** * **Master node**: Three configuration instances, 4 vCPUs and 12 GB RAM. * **System node**: Two configuration instances, 24 vCPUs and 64 GB RAM. * **Proxy node**: Two instances: One HTTP proxy and one RPC proxy, both with 4 vCPUs and 8 GB RAM. * Service **tablet node**: Two `c8-m16` configuration instances, 8 vCPUs and 16 GB RAM. * **Storage** with the following configuration: * Exec node storage: One non-replicated SSD, 93 GB. * Storage for service storage nodes: Three non-replicated SSDs, 93 GB each. * Storage for storage nodes: Six non-replicated SSDs, 93 GB each. {% cut "More info on service disks for storage nodes" %} 1. For three storage nodes for three non-replicated SSDs, 93 GB each, there will be three service disks, also non-replicated SSDs, 93 GB each. 1. For three storage nodes for three non-replicated SSDs, 465 GB each, there will be three service disks, also non-replicated SSDs, 93 GB each. {% endcut %} * **Storage node**: Six storage nodes, each with 8 vCPUs and 8 GB RAM; or three storage nodes, each with 4 vCPUs and 8 GB RAM. {% cut "More info on storage nodes" %} 1. For a storage with three non-replicated SSDs, 93 GB each, there are three service storage nodes, each with 8 vCPUs and 8 GB RAM. 1. For a storage with three HDDs, 2048 GB each, there are three storage nodes, each with 4 vCPUs and 8 GB RAM. 1. For a storage with three non-replicated SSDs, 465 GB each, there are three storage nodes, each with 8 vCPUs and 8 GB RAM. {% endcut %} #### Configurable components Cost calculation for an exec node: > 8 × $0.0113114736 + 32 × $0.0030245897 = $0.1872786592 > > Total: $0.1872786592, per hour cost of an exec node. Where: * 8: Number of vCPUs of one exec node. * $0.0113114736: Cost per vCPU per hour. * 32: Amount of RAM (GB) of one exec node. * $0.0030245897: Cost of using 1 GB of RAM per hour. Cost calculation for a tablet node: > 3 × (8 × $0.0113114736 + 16 × $0.0030245897) = $0.416655672 > > Total: $0.416655672, per hour cost of tablet nodes. Where: * 3: Number of tablet nodes. * 8: Number of vCPUs of one tablet node. * $0.0113114736: Cost per vCPU per hour. * 16: Amount of RAM (GB) of one tablet node. * $0.0030245897: Cost of using 1 GB of RAM per hour. Calculating the cost of cluster storage: > 3 × 2,048 × $0.0312786835 + 3 × 465 × $0.0950163782 = $324.724079013 > > Total: $324.724079013, cost of using non-replicated SSD and network HDD storage per month. Where: * 3: Number of HDDs. * 2,048: Size of one HDD (in GB). * $0.0312786835: Cost of using one GB of network HDD storage per month. * 3: Number of non-replicated SSDs. * 465: Size of one non-replicated SSD (in GB). * $0.0950163782: Cost of using one GB of non-replicated SSD storage per month. #### Service components Cost calculation for a master node: > 3 × (4 × $0.0113114736 + 12 × $0.0030245897) = $0.2446229124 > > Total: $0.2446229124, per hour cost of master nodes. Where: * 3: Number of master nodes. * 4: Number of vCPUs of one master node. * $0.0113114736: Cost per vCPU per hour. * 12: Amount of RAM (GB) of one master node. * $0.0030245897: Cost of using 1 GB of RAM per hour. Cost calculation for a system node: > 2 × (24 × $0.0113114736 + 64 × $0.0030245897) = $0.9300982144 > > Total: $0.9300982144, per hour cost of system nodes. Where: * 2: Number of system nodes. * 24: Number of vCPUs of one system node. * $0.0113114736: Cost per vCPU per hour. * 64: Amount of RAM (GB) of one system node. * $0.0030245897: Cost of using 1 GB of RAM per hour. Cost calculation for a proxy node: > 2 × (4 × $0.0113114736 + 8 × $0.0030245897) = $0.138885224 > > Total: $0.138885224, per hour cost of proxy nodes. Where: * 2: Number of proxy nodes. * 4: Number of vCPUs of one proxy node. * $0.0113114736: Cost per vCPU per hour. * 8: Amount of RAM (GB) of one proxy node. * $0.0030245897: Cost of using 1 GB of RAM per hour. Cost calculation for a service tablet node: > 2 × (8 × $0.0113114736 + 16 × $0.0030245897) = $0.277770448 > > Total: $0.277770448, per hour cost of service tablet nodes. Where: * 2: Number of tablet nodes. * 8: Number of vCPUs of one tablet node. * $0.0113114736: Cost per vCPU per hour. * 16: Amount of RAM (GB) of one tablet node. * $0.0030245897: Cost of using 1 GB of RAM per hour. Cost calculation for the exec node storage: > 93 × 1 × $0.0950163782 = $8.8365231726 > > Total: $8.8365231726, cost of using non-replicated SSD storage for the exec node per month. Where: * 93: Size of non-replicated SSD storage (in GB). * 1: Number of non-replicated SSDs. * $0.0950163782: Cost of using one GB of non-replicated SSD storage per month. Cost calculation for the service storage node storage: > 3 × 93 × $0.0950163782 = $26.5095695178 > > Total: $26.5095695178, cost of using non-replicated SSD storage for the service storage nodes per month. Where: * 3: Number of disks. * 93: Size of one non-replicated SSD (in GB). * $0.0950163782: Cost of using one GB of non-replicated SSD storage per month. Cost calculation for a storage node: > 6 × (8 × $0.0113114736 + 8 × $0.0030245897) + 3 × (4 × $0.0113114736 + 8 × $0.0030245897) = $0.8964588744 > > Total: $0.8964588744, per hour cost of storage nodes. Where: * 6: Number of storage nodes for SSDs or non-replicated SSDs. * 3: Number of storage nodes for HDDs. * 8: Number of vCPUs per storage node for SSDs or non-replicated SSDs. * 4: Number of vCPUs per storage node for HDDs. * $0.0113114736: Cost per vCPU per hour. * 8: Amount of RAM (GB) per storage node for SSDs or non-replicated SSDs. * $0.0030245897: Cost of using 1 GB of RAM per hour. Cost calculation for storage with service disks: > 93 × 6 × $0.0950163782 = $53.0191390356 > > Total: $53.0191390356, cost of using non-replicated SSD storage per month. Where: * 6: Number of service disks. * 93: Size of one non-replicated SSD (in GB). * $0.0950163782: Cost of using one GB of non-replicated SSD storage per month. #### Total cost calculation > 720 × ($0.1872786592 + $0.416655672 + $0.2446229124 + $0.9300982144 + $0.138885224 + $0.277770448 + $0.8964588744) + $324.724079013 + $8.8365231726 + $26.5095695178 + $53.0191390356 = $2,639.163713907 > > Total: $2,639.163713907, cost of using the cluster for 30 days. Where: * 720: Hours in 30 days. * $0.1872786592: Per hour cost of an exec node. * $0.416655672: Per hour cost of tablet nodes. * $0.2446229124: Per hour cost of master nodes. * $0.9300982144: Per hour cost of system nodes. * $0.138885224: Per hour cost of proxy nodes. * $0.277770448: Per hour cost of service tablet nodes. * $0.8964588744: Per hour cost of storage nodes. * $324.724079013: Cost of using non-replicated SSD and network HDD storage per month. * $8.8365231726: Cost of using non-replicated SSD storage for the exec node per month. * $26.5095695178: Cost of using non-replicated SSD storage for the service storage nodes per month. * $53.0191390356: Cost of using non-replicated SSD storage per month. ## Prices for the Russia region {#prices} {% note info %} Yandex Cloud resources are priced differently in different regions. Learn more about available regions in [Regions](../overview/concepts/region.md). Your payment currency is determined by your contracting legal entity. For more information on creating an account, see [Registering an account in Yandex Cloud](../billing/quickstart/index.md). {% endnote %} The monthly prices are based on 720 hours per month. | Услуга | Цена | Ед. тарификации | Действует с | Действует до | | ---------------------------------------------------------------------------------------------------------------------------------------------------- | ------------- | --------------- | -------------- | ------------ | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (AMD EPYC™ with NVIDIA® Ampere® A100), 1 GPU NVIDIA® Ampere® A100 | $4.01434362 | GPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (AMD EPYC™ with NVIDIA® Ampere® A100), RAM | $0.00302377 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (AMD EPYC™ with NVIDIA® Ampere® A100), vCPU | $0.0096721296 | vCPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (Intel Cascade Lake with NVIDIA® Tesla® V100), 1 GPU NVIDIA® Tesla® V100 | $2.570081556 | GPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (Intel Cascade Lake with NVIDIA® Tesla® V100), RAM | $0.0031926224 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (Intel Cascade Lake with NVIDIA® Tesla® V100), vCPU | $0.0122745882 | vCPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (Intel Ice Lake with T4i), 1 GPU | $1.5551636856 | GPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (Intel Ice Lake with T4i), RAM | $0.00302377 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (Intel Ice Lake with T4i), vCPU | $0.0113114736 | vCPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (Intel Ice Lake), RAM | $0.0030245897 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Exec Node (Intel Ice Lake), vCPU | $0.0113114736 | vCPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Master Node (Intel Ice Lake), RAM | $0.0030245897 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Master Node (Intel Ice Lake), vCPU | $0.0113114736 | vCPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Proxy Node (Intel Ice Lake), RAM | $0.0030245897 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Proxy Node (Intel Ice Lake), vCPU | $0.0113114736 | vCPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Storage Node (Intel Ice Lake), RAM | $0.0030245897 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Storage Node (Intel Ice Lake), vCPU | $0.0113114736 | vCPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, System Node (Intel Ice Lake), RAM | $0.0030245897 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, System Node (Intel Ice Lake), vCPU | $0.0113114736 | vCPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Tablet Node (Intel Ice Lake), RAM | $0.0030245897 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Вычислительные ресурсы компонентов, Tablet Node (Intel Ice Lake), vCPU | $0.0113114736 | vCPU × час | April 30, 2026 | — | | Managed Service for YTsaurus. Хранилище на нереплицируемых SSD-дисках | $0.0001319672 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Хранилище на сетевых HDD-дисках | $0.0000434426 | ГБ × час | April 30, 2026 | — | | Managed Service for YTsaurus. Хранилище на сетевых SSD-дисках | $0.0001778688 | ГБ × час | April 30, 2026 | — | ### Outgoing traffic {#prices-traffic} When using the service, you pay for traffic from Yandex Cloud to the internet. Traffic between internal IP addresses of Yandex Cloud services and incoming internet traffic is free. The first 100 GB of outgoing traffic per month are free of charge. The minimum billing unit is 1 MB. | Услуга | Цена | Ед. тарификации | Действует с | Действует до | | --------------------------------------------------------- | ----------------- | --------------- | -------------- | ------------ | | Исходящий трафик, от 0 до 100 единицы тарификации в месяц | Не тарифицируется | ГБ | April 30, 2026 | — | | Исходящий трафик, от 100 единицы тарификации в месяц | $0.0116393424 | ГБ | April 30, 2026 | — |