{"id":16911,"date":"2025-06-17T15:58:45","date_gmt":"2025-06-17T14:58:45","guid":{"rendered":"https:\/\/ibertronica.es\/blog\/?p=16911"},"modified":"2025-06-17T15:58:45","modified_gmt":"2025-06-17T14:58:45","slug":"the-technology-behind-the-cloud-for-ai","status":"publish","type":"post","link":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/","title":{"rendered":"The technology behind the Cloud for AI"},"content":{"rendered":"<p><span data-contrast=\"auto\">Current generative AI models\u2014capable of handling hundreds of billions of parameters\u2014have exponentially increased the computing demands on data centers and the corporate clouds that support them (Cloud for AI). Training and inferring these models without bottlenecks requires moving terabytes of data per second, dissipating more than 100kW per rack, and scaling computing capacity efficiently. This challenge can only be met by relying on four well-defined technological pillars:<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<ul>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1001\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"0\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Latest generation GPUs<\/span><\/b><span data-contrast=\"auto\">, with HBM memory and NVLink\/NVSwitch links.<\/span><\/li>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1001\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"0\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Ultra-dense servers<\/span><\/b><span data-contrast=\"auto\">, which maximize the use of each rack unit.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:36,&quot;335559739&quot;:36}\">\u00a0<\/span><\/li>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1001\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"0\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Advanced liquid cooling<\/span><\/b><span data-contrast=\"auto\">, capable of operating with warm water to extract high thermal loads.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:36,&quot;335559739&quot;:36}\">\u00a0<\/span><\/li>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1001\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"0\" data-aria-level=\"1\"><b><b> This type of <strong data-start=\"621\" data-end=\"638\">AI cloud<\/strong> requires complete integration between specialized hardware, efficient cooling, and high-speed connectivity to support next-generation models.<\/b><\/b><a onclick=\"javascript:pageTracker._trackPageview('\/downloads\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Diagrama.jpg');\"  href=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Diagrama.jpg\"><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter wp-image-16883 size-full\" title=\"Tecnolog\u00eda Cloud Para Ia Diagrama\" src=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Diagrama.jpg\" alt=\"Tecnolog\u00eda Cloud Para Ia Diagrama\" width=\"1024\" height=\"1024\" srcset=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Diagrama.jpg 1024w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Diagrama-300x300.jpg 300w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Diagrama-150x150.jpg 150w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Diagrama-116x116.jpg 116w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/a><span data-contrast=\"auto\">Throughout the article, we break down each of these pillars, explaining how they combine in practice and what real technologies make them possible.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><br \/>\n<h2 aria-level=\"3\"><span data-contrast=\"none\">GPUs: The Engine of AI<\/span><span data-ccp-props=\"{&quot;134245418&quot;:true,&quot;134245529&quot;:true,&quot;335559738&quot;:160,&quot;335559739&quot;:80}\">\u00a0<\/span><\/h2>\n<p><span data-contrast=\"auto\">Graphics processors lead the way in computing performance today thanks to their massive ability to perform floating-point operations (<\/span><b><span data-contrast=\"auto\">FLOPS<\/span><\/b><span data-contrast=\"auto\">) and move data at high speed. A single modern GPU offers up to 288GB of HBM3e memory and over 8TB\/s of internal bandwidth, figures impossible to achieve with any conventional CPU.<\/span><\/p>\n<p>Cards like the NVIDIA H100\/H200 or the new AMD Instinct MI350 data-contrast=\u00bbauto\u00bb&gt; combine this memory with tens of thousands of cores and high-speed links like<\/p>\n<p><b><b><b><span data-contrast=\"auto\">NVLink<\/span><\/b><span data-contrast=\"auto\"> and <\/span><b><span data-contrast=\"auto\">NVSwitch<\/span><\/b><span data-contrast=\"auto\">, which allow multiple GPUs to function as one. This frees up traditional CPUs for coordination tasks, while the GPUs focus on intensive computation.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/b><\/b>Manufacturers such as ASUS and Gigabyte have already adapted their servers to take advantage of these capabilities. For example, nodes with up to eight GPUs interconnected by NVSwitch can deliver more than <b><span data-contrast=\"auto\">4PFLOPS<\/span><\/b><span data-contrast=\"auto\"> in FP16 precision, which is 30 times the power of a conventional CPU server rack.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><br \/>\n<span data-contrast=\"auto\">To take advantage of all this performance, the software must follow: libraries such as <\/span><b><span data-contrast=\"auto\">NCCL<\/span><\/b><span data-contrast=\"auto\">, <\/span><b><span data-contrast=\"auto\">GPUDirectRDMA<\/span><\/b><span data-contrast=\"auto\"> and <\/span><b><span data-contrast=\"auto\">GPUDirectStorage<\/span><\/b><span data-contrast=\"auto\"> allow data to flow between GPUs, network and storage without going through the system RAM, reducing latencies to the microsecond level.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<h2 aria-level=\"3\">Dense Cloud Servers for AI: More Power in Less Space<\/h2>\n<p><span data-contrast=\"auto\">The key metric in an AI cluster is not just raw power, but how much of that power fits in each rack (<\/span><b><span data-contrast=\"auto\">FLOPS\/U<\/span><\/b><span data-contrast=\"auto\">). To achieve extreme densities, servers with multiple high-performance GPUs are clustered together, precisely positioned to maximize space utilization, cooling, and internal connectivity.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<p><span data-contrast=\"auto\">A well-optimized 42U rack might include:<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<ul>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1002\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"4\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">4\u20136 8GPU servers<\/span><\/b><span data-contrast=\"auto\">, interconnected via PCIe Gen5 buses or NVSwitch links.<\/span><\/li>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1002\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"4\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Network and management switches<\/span><\/b><span data-contrast=\"auto\"> (200Gb\/s), top mounted.<\/span><\/li>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1002\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"4\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Distributed power modules (48\u201354VDC busbar, exclusive to OCPV3 racks) and, depending on the overall design, also thermal control modules or external racks dedicated to shared cooling.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:36,&quot;335559739&quot;:36}\">\u00a0<\/span><\/b><\/li>\n<\/ul>\n<p><span data-contrast=\"auto\">This type of architecture can achieve <\/span><b><span data-contrast=\"auto\">20\u201325 FP16 PFLOPS per rack<\/span><\/b><span data-contrast=\"auto\">. However, the final performance will also depend on the <\/span><i><span data-contrast=\"auto\">communications software<\/span><\/i><span data-contrast=\"auto\">, which must be adapted to the GPU manufacturer:<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<ul>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1003\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"7\" data-aria-level=\"1\"><span data-contrast=\"auto\">With <\/span><b><span data-contrast=\"auto\">NVIDIA<\/span><\/b><span data-contrast=\"auto\">: libraries <\/span><b><span data-contrast=\"auto\">NCCL<\/span><\/b><span data-contrast=\"auto\"> and <\/span><b><span data-contrast=\"auto\">GPUDirectRDMA<\/span><\/b><span data-contrast=\"auto\">.<\/span><\/li>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1003\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"7\" data-aria-level=\"1\"><span data-contrast=\"auto\">With <\/span><b><span data-contrast=\"auto\">AMD<\/span><\/b><span data-contrast=\"auto\">: using <\/span><b><span data-contrast=\"auto\">RCCL<\/span><\/b><span data-contrast=\"auto\"> and <\/span><b><span data-contrast=\"auto\">ROCmDirectRDMA<\/span><\/b><span data-contrast=\"auto\">.<\/span><\/li>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1003\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"7\" data-aria-level=\"1\"><span data-contrast=\"auto\">With <\/span><b><span data-contrast=\"auto\">Intel<\/span><\/b><span data-contrast=\"auto\">: libraries <\/span><b><span data-contrast=\"auto\">oneCCL<\/span><\/b><span data-contrast=\"auto\"> with support in <\/span><b><span data-contrast=\"auto\">oneAPI\/LevelZero<\/span><\/b><span data-contrast=\"auto\">.<\/span><\/li>\n<\/ul>\n<p><span data-contrast=\"auto\">Each ecosystem has its own optimized stack, so it\u2019s a good idea to define the hardware platform before choosing the software environment.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<h2 aria-level=\"3\">Cloud Cooling for AI: Air or Liquid?<\/h2>\n<p><span data-contrast=\"auto\">Before designing a cluster, you need to decide how to dissipate the <\/span><b><span data-contrast=\"auto\">20\u201340kW<\/span><\/b><span data-contrast=\"auto\"> that an AI rack can generate. There are two main approaches:<\/span><\/p>\n<ul>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1004\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"10\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Air cooling<\/span><\/b><span data-contrast=\"auto\">: This is the simplest and uses high-flow fans. It&#8217;s usually sufficient for small installations (up to three racks).<\/span><span data-ccp-props=\"{&quot;335559738&quot;:36,&quot;335559739&quot;:36}\">\u00a0<\/span><\/li>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1004\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"10\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Liquid cooling (LC)<\/span><\/b><span data-contrast=\"auto\">: more efficient starting at 20kW\/rack. It uses glycol water to extract heat directly from the GPUs and CPUs using <\/span><i><span data-contrast=\"auto\">cold-plates<\/span><\/i><span data-contrast=\"auto\">.<\/span><\/li>\n<\/ul>\n<p><span data-contrast=\"auto\">Within LC there are two modalities:<\/span><\/p>\n<ul>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1005\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"12\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Liquid-to-Air (L2A)<\/span><\/b><span data-contrast=\"auto\">: Hot water is cooled in an exchanger inside the rack, dissipating the heat as warm air.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:36,&quot;335559739&quot;:36}\">\u00a0<\/span><\/li>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1005\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"12\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Liquid-to-Liquid (L2L)<\/span><\/b><span data-contrast=\"auto\">: The water transfers its heat to an external circuit, completely removing the heat from the technical room.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:36,&quot;335559739&quot;:36}\">\u00a0<\/span><\/li>\n<\/ul>\n<p><span data-contrast=\"auto\">These solutions are not mutually exclusive. In many cases, they are combined: for example, liquid cooling for GPUs and air cooling for disks and secondary components.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<p><span data-contrast=\"auto\">The LC can be integrated into the rack itself (<\/span><b><span data-contrast=\"auto\">in-rack CDU<\/span><\/b><span data-contrast=\"auto\">) or managed from an external cabinet serving multiple racks (<\/span><b><span data-contrast=\"auto\">remote CDU<\/span><\/b><span data-contrast=\"auto\">).<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<p><b><span data-contrast=\"auto\">Important Note:<\/span><\/b><span data-contrast=\"auto\">NVIDIA Blackwell (B100\/B200) generation<\/span><\/p>\n<p><span data-contrast=\"auto\"> requires liquid cooling as it exceeds 1kW per GPU. Previous generations (like the H100\/H200) can still work over the air if you have a well-conditioned room.<\/span><\/p>\n<h2 aria-level=\"3\"><span data-contrast=\"none\">200Gb\/s Networks: Moving Data Without Bottlenecks<\/span><span data-ccp-props=\"{&quot;134245418&quot;:true,&quot;134245529&quot;:true,&quot;335559738&quot;:160,&quot;335559739&quot;:80}\">\u00a0<\/span><\/h2>\n<p><span data-contrast=\"auto\">To train large models, each GPU can send or receive up to <\/span><b><span data-contrast=\"auto\">50\u2013100GB per iteration<\/span><\/b><span data-contrast=\"auto\">. On a 32-node cluster, this generates several <\/span><b><span data-contrast=\"auto\">terabits per second<\/span><\/b><span data-contrast=\"auto\"> of traffic. To avoid bottlenecks, each GPU should have its own 200Gb\/s network-attached NIC, connected to the network and aligned with its NUMA domain.<\/span><\/p>\n<h3><b><span data-contrast=\"auto\">How the Rack Network Is Structured in an AI Cloud<\/span><\/b><\/h3>\n<ol>\n<li data-leveltext=\"%1.\" data-font=\"\" data-listid=\"1006\" data-list-defn-props=\"{&quot;335552541&quot;:0,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769242&quot;:[65533 ,0],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;%1.&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"1\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">GPU\u2013NIC connection<\/span><\/b><span data-contrast=\"auto\">: Each GPU connects to its own 200Gb\/s NIC (InfiniBand or Ethernet).<\/span><\/li>\n<li data-leveltext=\"%1.\" data-font=\"\" data-listid=\"1006\" data-list-defn-props=\"{&quot;335552541&quot;:0,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769242&quot;:[65533 ,0],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;%1.&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"1\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Switch leaf<\/span><\/b><span data-contrast=\"auto\">: The NICs connect to switches at the top of the rack. These leaf switches act as the entry point to the cluster network. <\/span><b><span data-contrast=\"auto\">\u00a0<\/span><\/b><span data-contrast=\"auto\"><br \/>\n<\/span><\/li>\n<li data-leveltext=\"%1.\" data-font=\"\" data-listid=\"1006\" data-list-defn-props=\"{&quot;335552541&quot;:0,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769242&quot;:[65533 ,0],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;%1.&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"1\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Switch spine<\/span><\/b><span data-contrast=\"auto\">: Each leaf connects to several <\/span><b><span data-contrast=\"auto\">spines<\/span><\/b><span data-contrast=\"auto\">, backbone switches that interconnect the different leafs. Thus, any GPU can communicate with another in a maximum of two hops: GPU \u2192 leaf \u2192 spine \u2192 leaf \u2192 GPU.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:36,&quot;335559739&quot;:36}\">\u00a0<\/span><\/li>\n<li data-leveltext=\"%1.\" data-font=\"\" data-listid=\"1006\" data-list-defn-props=\"{&quot;335552541&quot;:0,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769242&quot;:[65533 ,0],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;%1.&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"1\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Scaling<\/span><\/b><span data-contrast=\"auto\">: This leaf-and-spine topology works well for up to 256 nodes. For larger deployments, move to more complex structures such as <\/span><b><span data-contrast=\"auto\">dragonfly+<\/span><\/b><span data-contrast=\"auto\">.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:36,&quot;335559739&quot;:36}\">\u00a0<\/span><\/li>\n<\/ol>\n<h3><b><span data-contrast=\"auto\">Rule of Thumb<\/span><\/b><span data-contrast=\"auto\">:<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/h3>\n<p><span data-contrast=\"auto\">The total network bandwidth of your rack should be similar to the aggregate memory bandwidth of your GPUs, so as not to waste computing power.<\/span><\/p>\n<h4><b><span data-contrast=\"auto\">Hardware examples (200Gb\/s)<\/span><\/b><span data-contrast=\"auto\">:<\/span><\/h4>\n<ul>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1007\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"14\" data-aria-level=\"1\"><span data-contrast=\"auto\">NICs: NVIDIA ConnectX\u20117, AMD Pensando DPU, Intel E810\u2011CQDA2<\/span><\/li>\n<li data-leveltext=\"\uf0b7\" data-font=\"Symbol\" data-listid=\"1007\" data-list-defn-props=\"{&quot;335551671&quot;:0,&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769226&quot;:&quot;Symbol &quot;,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\uf0b7&quot;,&quot;469777815&quot;:&quot;multilevel&quot;}\" aria-setsize=\"-1\" data-aria-posinset=\"14\" data-aria-level=\"1\"><span data-contrast=\"auto\">Switches: NVIDIA InfiniBand QM9700, Broadcom Tomahawk6 (Ethernet)<\/span><span data-ccp-props=\"{&quot;335559738&quot;:36,&quot;335559739&quot;:36}\">\u00a0<\/span><\/li>\n<\/ul>\n<p><span data-contrast=\"auto\">These components support <\/span><b><span data-contrast=\"auto\">RDMA<\/span><\/b><span data-contrast=\"auto\">, a technology that allows GPUs to exchange data directly without going through the CPU, improving efficiency and reducing latency.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter wp-image-16884 size-full\" title=\"Cloud Technology for the Network\" src=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Red.jpg\" alt=\"Cloud Technology for the Network\" width=\"800\" height=\"800\" srcset=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Red.jpg 800w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Red-300x300.jpg 300w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Red-150x150.jpg 150w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Tecnologia-cloud-para-IA-Red-116x116.jpg 116w\" sizes=\"auto, (max-width: 800px) 100vw, 800px\" \/><\/p>\n<h3 aria-level=\"3\"><span data-contrast=\"none\">Conclusion: accelerate AI without improvising<\/span><span data-ccp-props=\"{&quot;134245418&quot;:true,&quot;134245529&quot;:true,&quot;335559738&quot;:160,&quot;335559739&quot;:80}\">\u00a0<\/span><\/h3>\n<p>Adopting advanced AI means transforming the infrastructure. It&#8217;s not enough to simply buy powerful servers: you have to design a coherent AI cloud, where computing, cooling, and networking work in harmony.<br \/>\n<span data-contrast=\"auto\">GPU servers from ASUS, Gigabyte, ASRock Rack, and Supermicro demonstrate how the ecosystem is already prepared to deliver solutions ready to train large-scale models. But deploying them successfully requires experience, precision, and practical knowledge.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<p><b><span data-contrast=\"auto\">ViveRack<\/span><\/b><span data-contrast=\"auto\">, the line of high-performance solutions marketed by <\/span><b><span data-contrast=\"auto\">ibertronica<\/span><\/b><span data-contrast=\"auto\">, brings together all of these elements in proven configurations: certified GPU nodes, cooling distribution, 200Gb\/s switching, and an architecture that is ready to grow.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<p><span data-contrast=\"auto\">\ud83d\udc49 If your company is evaluating a qualitative leap in AI, we invite you to learn more about ViveRack configurations and discover how we can help you turn your data center into a true model factory.<\/span><span data-ccp-props=\"{&quot;335559738&quot;:180,&quot;335559739&quot;:180}\">\u00a0<\/span><\/p>\n<p><b><span data-contrast=\"auto\">You can contact us through the <a href=\"https:\/\/ibertronica.es\/contact\">form<\/a> available on our website to receive personalized advice.<\/span><\/b><\/p>\n<table style=\"border-collapse: collapse; width: 100%;\">\n<tbody>\n<tr>\n<td style=\"width: 33.3333%; text-align: center;\"><strong>How to Deploy Your Cloud for AI With VibeRack Racks<\/strong><\/td>\n<td style=\"width: 33.3333%; text-align: center;\"><strong>New Ocp V3 Cabinets<\/strong><\/td>\n<td style=\"width: 33.3333%; text-align: center;\"><strong>NVIDIA HGX: Open Platform Powering AI and HPC at Large Scale<\/strong><\/td>\n<\/tr>\n<tr>\n<td style=\"width: 33.3333%;\"><a href=\"https:\/\/ibertronica.es\/blog\/actualidad\/como-desplegar-tu-cloud-para-ia-con-racks-viberack-de-ibertronica\/\"><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter wp-image-16889 size-full\" title=\"Cabecera Guia Cloud Para Ia\" src=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Guia-Cloud-para-IA.jpg\" alt=\"Cabecera Guia Cloud Para Ia\" width=\"1000\" height=\"571\" srcset=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Guia-Cloud-para-IA.jpg 1000w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Guia-Cloud-para-IA-300x171.jpg 300w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Guia-Cloud-para-IA-150x86.jpg 150w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Guia-Cloud-para-IA-203x116.jpg 203w\" sizes=\"auto, (max-width: 1000px) 100vw, 1000px\" \/><\/a><\/td>\n<td style=\"width: 33.3333%;\"><a href=\"https:\/\/ibertronica.es\/blog\/actualidad\/nuevos-armarios-ocp-v3-cloud-para-ia\/\"><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter wp-image-16900 size-full\" title=\"Cabecera Nuevos Armarios Ocp V3 2\" src=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Nuevos-Armarios-Ocp-V3-2.jpg\" alt=\"Cabecera Nuevos Armarios Ocp V3 2\" width=\"1000\" height=\"571\" srcset=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Nuevos-Armarios-Ocp-V3-2.jpg 1000w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Nuevos-Armarios-Ocp-V3-2-300x171.jpg 300w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Nuevos-Armarios-Ocp-V3-2-150x86.jpg 150w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Nuevos-Armarios-Ocp-V3-2-203x116.jpg 203w\" sizes=\"auto, (max-width: 1000px) 100vw, 1000px\" \/><\/a><\/td>\n<td style=\"width: 33.3333%;\"><a href=\"https:\/\/ibertronica.es\/blog\/actualidad\/nvidia-hgx-plataforma-abierta-que-impulsa-la-ia-y-hpc-a-gran-escala\/\"><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter wp-image-16906 size-full\" title=\"Cabecera Nvidia Hgx\" src=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-nvidia-hgx.jpg\" alt=\"Cabecera Nvidia Hgx\" width=\"1000\" height=\"571\" srcset=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-nvidia-hgx.jpg 1000w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-nvidia-hgx-300x171.jpg 300w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-nvidia-hgx-150x86.jpg 150w, https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-nvidia-hgx-203x116.jpg 203w\" sizes=\"auto, (max-width: 1000px) 100vw, 1000px\" \/><\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Current generative AI models\u2014capable of handling hundreds of billions of parameters\u2014have exponentially increased the computing demands on data centers and the corporate clouds that support them (Cloud for AI). Training and inferring these models without bottlenecks requires moving terabytes of data per second, dissipating more than 100kW per rack, and scaling computing capacity efficiently. This challenge can only be met by relying on four well-defined technological pillars:\u00a0 Latest generation GPUs, with HBM memory and NVLink\/NVSwitch links. Ultra-dense servers, which maximize&hellip;<\/p>\n","protected":false},"author":2,"featured_media":16882,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1410],"tags":[4856],"class_list":["post-16911","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news-en","tag-cloud-for-ai","post-has-thumbnail"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.1 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Cloud for AI: GPUs, Network, and Cooling<\/title>\n<meta name=\"description\" content=\"Cloud infrastructure for AI with high-performance GPUs, liquid cooling, and 200 Gb\/s networks to train advanced models.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/\" \/>\n<meta property=\"og:locale\" content=\"es_ES\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Cloud for AI: GPUs, Network, and Cooling\" \/>\n<meta property=\"og:description\" content=\"Cloud infrastructure for AI with high-performance GPUs, liquid cooling, and 200 Gb\/s networks to train advanced models.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/\" \/>\n<meta property=\"og:site_name\" content=\"Blog de tecnolog\u00eda\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/IbertronicaES\/\" \/>\n<meta property=\"article:published_time\" content=\"2025-06-17T14:58:45+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Tecnologia-cloud-para-IA.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"571\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"magazine\" \/>\n<meta name=\"twitter:label1\" content=\"Escrito por\" \/>\n\t<meta name=\"twitter:data1\" content=\"magazine\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tiempo de lectura\" \/>\n\t<meta name=\"twitter:data2\" content=\"6 minutos\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/\"},\"author\":{\"name\":\"magazine\",\"@id\":\"https:\/\/ibertronica.es\/blog\/#\/schema\/person\/0375a85dbad874ed61652f20a07ea33f\"},\"headline\":\"The technology behind the Cloud for AI\",\"datePublished\":\"2025-06-17T14:58:45+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/\"},\"wordCount\":1144,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/ibertronica.es\/blog\/#organization\"},\"image\":{\"@id\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Tecnologia-cloud-para-IA.jpg\",\"keywords\":[\"cloud for AI\"],\"articleSection\":[\"News\"],\"inLanguage\":\"es\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/\",\"url\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/\",\"name\":\"Cloud for AI: GPUs, Network, and Cooling\",\"isPartOf\":{\"@id\":\"https:\/\/ibertronica.es\/blog\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Tecnologia-cloud-para-IA.jpg\",\"datePublished\":\"2025-06-17T14:58:45+00:00\",\"description\":\"Cloud infrastructure for AI with high-performance GPUs, liquid cooling, and 200 Gb\/s networks to train advanced models.\",\"breadcrumb\":{\"@id\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#breadcrumb\"},\"inLanguage\":\"es\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"es\",\"@id\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#primaryimage\",\"url\":\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Tecnologia-cloud-para-IA.jpg\",\"contentUrl\":\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Tecnologia-cloud-para-IA.jpg\",\"width\":1000,\"height\":571,\"caption\":\"Tecnolog\u00eda Cloud Para Ia\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Portada\",\"item\":\"https:\/\/ibertronica.es\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"The technology behind the Cloud for AI\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/ibertronica.es\/blog\/#website\",\"url\":\"https:\/\/ibertronica.es\/blog\/\",\"name\":\"Blog de tecnolog\u00eda\",\"description\":\"Ibertr\u00f3nica, un blog sobre hardware inform\u00e1tico y servidores para todo tipo de prestaciones\",\"publisher\":{\"@id\":\"https:\/\/ibertronica.es\/blog\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/ibertronica.es\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"es\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/ibertronica.es\/blog\/#organization\",\"name\":\"Sistemas Ibertr\u00f3nica\",\"url\":\"https:\/\/ibertronica.es\/blog\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"es\",\"@id\":\"https:\/\/ibertronica.es\/blog\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2020\/03\/logotipo_web_2019-1.png\",\"contentUrl\":\"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2020\/03\/logotipo_web_2019-1.png\",\"width\":417,\"height\":45,\"caption\":\"Sistemas Ibertr\u00f3nica\"},\"image\":{\"@id\":\"https:\/\/ibertronica.es\/blog\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/IbertronicaES\/\",\"https:\/\/x.com\/Ibertronica_Es\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/ibertronica.es\/blog\/#\/schema\/person\/0375a85dbad874ed61652f20a07ea33f\",\"name\":\"magazine\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"es\",\"@id\":\"https:\/\/ibertronica.es\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/7d1ae06745040f270e4f310dafae96f9bc6d960c810d653ac06972e979569589?s=96&d=https%3A%2F%2Fwww.ibertronica.es%2Fimages%2Fperfil-azul.jpg&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/7d1ae06745040f270e4f310dafae96f9bc6d960c810d653ac06972e979569589?s=96&d=https%3A%2F%2Fwww.ibertronica.es%2Fimages%2Fperfil-azul.jpg&r=g\",\"caption\":\"magazine\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Cloud for AI: GPUs, Network, and Cooling","description":"Cloud infrastructure for AI with high-performance GPUs, liquid cooling, and 200 Gb\/s networks to train advanced models.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/","og_locale":"es_ES","og_type":"article","og_title":"Cloud for AI: GPUs, Network, and Cooling","og_description":"Cloud infrastructure for AI with high-performance GPUs, liquid cooling, and 200 Gb\/s networks to train advanced models.","og_url":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/","og_site_name":"Blog de tecnolog\u00eda","article_publisher":"https:\/\/www.facebook.com\/IbertronicaES\/","article_published_time":"2025-06-17T14:58:45+00:00","og_image":[{"width":1000,"height":571,"url":"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Tecnologia-cloud-para-IA.jpg","type":"image\/jpeg"}],"author":"magazine","twitter_misc":{"Escrito por":"magazine","Tiempo de lectura":"6 minutos"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#article","isPartOf":{"@id":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/"},"author":{"name":"magazine","@id":"https:\/\/ibertronica.es\/blog\/#\/schema\/person\/0375a85dbad874ed61652f20a07ea33f"},"headline":"The technology behind the Cloud for AI","datePublished":"2025-06-17T14:58:45+00:00","mainEntityOfPage":{"@id":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/"},"wordCount":1144,"commentCount":0,"publisher":{"@id":"https:\/\/ibertronica.es\/blog\/#organization"},"image":{"@id":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#primaryimage"},"thumbnailUrl":"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Tecnologia-cloud-para-IA.jpg","keywords":["cloud for AI"],"articleSection":["News"],"inLanguage":"es","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/","url":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/","name":"Cloud for AI: GPUs, Network, and Cooling","isPartOf":{"@id":"https:\/\/ibertronica.es\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#primaryimage"},"image":{"@id":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#primaryimage"},"thumbnailUrl":"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Tecnologia-cloud-para-IA.jpg","datePublished":"2025-06-17T14:58:45+00:00","description":"Cloud infrastructure for AI with high-performance GPUs, liquid cooling, and 200 Gb\/s networks to train advanced models.","breadcrumb":{"@id":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#breadcrumb"},"inLanguage":"es","potentialAction":[{"@type":"ReadAction","target":["https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/"]}]},{"@type":"ImageObject","inLanguage":"es","@id":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#primaryimage","url":"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Tecnologia-cloud-para-IA.jpg","contentUrl":"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2025\/06\/Cabecera-Tecnologia-cloud-para-IA.jpg","width":1000,"height":571,"caption":"Tecnolog\u00eda Cloud Para Ia"},{"@type":"BreadcrumbList","@id":"https:\/\/ibertronica.es\/blog\/en\/news-en\/the-technology-behind-the-cloud-for-ai\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Portada","item":"https:\/\/ibertronica.es\/blog\/"},{"@type":"ListItem","position":2,"name":"The technology behind the Cloud for AI"}]},{"@type":"WebSite","@id":"https:\/\/ibertronica.es\/blog\/#website","url":"https:\/\/ibertronica.es\/blog\/","name":"Blog de tecnolog\u00eda","description":"Ibertr\u00f3nica, un blog sobre hardware inform\u00e1tico y servidores para todo tipo de prestaciones","publisher":{"@id":"https:\/\/ibertronica.es\/blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/ibertronica.es\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"es"},{"@type":"Organization","@id":"https:\/\/ibertronica.es\/blog\/#organization","name":"Sistemas Ibertr\u00f3nica","url":"https:\/\/ibertronica.es\/blog\/","logo":{"@type":"ImageObject","inLanguage":"es","@id":"https:\/\/ibertronica.es\/blog\/#\/schema\/logo\/image\/","url":"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2020\/03\/logotipo_web_2019-1.png","contentUrl":"https:\/\/ibertronica.es\/blog\/wp-content\/uploads\/2020\/03\/logotipo_web_2019-1.png","width":417,"height":45,"caption":"Sistemas Ibertr\u00f3nica"},"image":{"@id":"https:\/\/ibertronica.es\/blog\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/IbertronicaES\/","https:\/\/x.com\/Ibertronica_Es"]},{"@type":"Person","@id":"https:\/\/ibertronica.es\/blog\/#\/schema\/person\/0375a85dbad874ed61652f20a07ea33f","name":"magazine","image":{"@type":"ImageObject","inLanguage":"es","@id":"https:\/\/ibertronica.es\/blog\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/7d1ae06745040f270e4f310dafae96f9bc6d960c810d653ac06972e979569589?s=96&d=https%3A%2F%2Fwww.ibertronica.es%2Fimages%2Fperfil-azul.jpg&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/7d1ae06745040f270e4f310dafae96f9bc6d960c810d653ac06972e979569589?s=96&d=https%3A%2F%2Fwww.ibertronica.es%2Fimages%2Fperfil-azul.jpg&r=g","caption":"magazine"}}]}},"_links":{"self":[{"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/posts\/16911","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/comments?post=16911"}],"version-history":[{"count":1,"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/posts\/16911\/revisions"}],"predecessor-version":[{"id":16912,"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/posts\/16911\/revisions\/16912"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/media\/16882"}],"wp:attachment":[{"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/media?parent=16911"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/categories?post=16911"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/ibertronica.es\/blog\/wp-json\/wp\/v2\/tags?post=16911"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}