Recent Content
Is Compute Scarcity Stalling Your AI Progress? ⚡
AI success isn’t just about the model. In the article, "Managing AI In An Era Of Compute Scarcity: Governance Takes Center Stage", Palanivel Rajan Mylsamy: Director of Engineering Program Management at Cisco explains why compute governance is the new priority for tech leaders. Mylsamy highlights that compute power has become the "new gold," making it the primary bottleneck for scaling. To succeed, organizations must move away from wasteful resource allocation and embrace intelligent routing and hybrid infrastructures that balance security with cost-effectiveness. Ultimately, you can’t scale AI on a weak foundation; true growth requires moving from "what’s possible" to "what’s operationally sound." 📣 Community Question: How is your organization handling the compute crunch? Are you leaning more toward cloud, on-prem or a hybrid model? Explain why. Let's discuss! Click through to read the entire article above and let us know your thoughts around it in the comments below!5Views0likes0CommentsEverything Ransomware: Ransomware Live
Check it out! Really interesting tracking of everything related to ransomware. https://www.ransomware.live/ Ransomware Live is a real-time intelligence site tracking active ransomware groups, victims, leaks, and extortion activity, helping security teams monitor threats, trends, and attacker behavior worldwide.NVMe-TCP + VMware Design Questions
Q1: In a medium - large datacenter (ie anticipated growth to 500 hosts), would it be a problem to use large subnets (ie 2 x /23 or /22's) to place the initiators + target IPs? All our existing nvme-tcp deployments (which we're having great success with btw) use 2 x /24's, which provides more than enough capacity for the array + a decent sized cluster. But now we're entering into the realms of multiple flasharrays (ie 3 or 4 in a single site) and hundreds of hosts - we'll quickly exhaust a /24 range. Q2: In this site, where we have say 2 x FA-X's an 2 x FA-C's, is it valid to present datastores from all of these arrays over nvme-tcp to a single cluster? With each array / datastore cluster being accessed using the same initiators on the same layer 2's? At what point does it get silly and I need to draw a line and segregate environments? Q3: Is there a capacity limit to the number of client hosts I can have on an array / mapping storage to? (nvm - I think I found the answer to that - looks like it's 1000 nqn's on pretty much all X and C arrays)191Views1like5CommentsPure Report Podcast: Nutanix and Pure Storage: Propelling Enterprise Virtualization Forward
Check out the latest edition of the Pure Report podcast where we unpack the GA announcement for the Nutanix and Pure Storage partnership. Hear from Cody_Hosterman and Nutanix VP Product Ketan Shah on the technical details of the integration and how this partnership came together.
15Views1like0CommentsNutanix and Pure Storage are Changing Virtualization
Big news, virtualization fans! The combined Nutanix + Pure Storage solution is now available. You can read all about it in the blog and get further details on our Nutanix partner page. We’ve been talking to lots of folks about this offering, both Pure users and not, and the consensus is that people are glad to see some new stability being brought to the virtualization world, with a solution from two customer-centric organizations. To give you a sense of the value of using external storage with FlashArray, an early adopter (I’m not at liberty to name them) running a nearly 2 PB database workload will save about 50% on rack space with significant savings on power, cooling and operational costs. Please contact your Pure sales team if you want to learn more about this solution.Proxmox VE
Hi all Hope you're all having a great day. We have several customers going down the Proxmox VE road. One of my colleagues was put onto https://github.com/kolesa-team/pve-purestorage-plugin as a possible solution (as using Pure behind Proxmox (using the native Proxmox release) is not a particularly Pure-like experience. Could someone from Pure comment on the plugin's validity/supportability?481Views2likes3CommentsJoin us for Nutanix IT Unplugged with Isaac Slade
Our friends at Nutanix are sponsering a fun virtual event on Dec. 5 for IT Unplugged featuring a live performance by Isaac Slade, formerly of The Fray. You'll hear from both Nutanix and Pure about our new joint solution, along with some musical fun. Go here for more details and to register: https://event.nutanix.com/itunplugged-december2025?utm_source=PureStorage22Views0likes0CommentsRansomware attacks are NOT going away
Here is why ransomware attacks are persistent and unlikely to disappear: 1. High Profitability and Low Risk for Criminals Ransomware is fundamentally a business model for organized crime, and it is overwhelmingly successful and profitable. Low Barrier to Entry: The rise of Ransomware-as-a-Service (RaaS) means even novice criminals can purchase sophisticated malware and infrastructure. This franchise model ensures high attack volume regardless of law enforcement efforts. Guaranteed Revenue Stream: The evolution to multi-extortion (encrypting data and stealing it) ensures that victims are forced to pay—either to regain system access or to prevent catastrophic data leaks and regulatory fines. This dual leverage guarantees profit even if the victim has backups. Anonymity: The use of cryptocurrency for payments, coupled with geopolitical safe zones for many RaaS groups, keeps the risk of prosecution extremely low for the attackers. 2. Attackers Are Outpacing Traditional Defenses The tactics used by ransomware groups are specifically designed to neutralize traditional defense and recovery measures: Targeting the Supply Chain: Attackers are finding success by targeting trusted vendors and IT providers to compromise dozens of companies simultaneously, making defense exponentially harder for individual organizations. Attacking Backups: Modern ransomware campaigns specifically target accessible backups to delete them or malware-infect them, eliminating the victim’s recovery option and forcing them to pay the ransom. AI for Stealth and Speed: The adoption of AI is accelerating reconnaissance and stealth, dramatically compressing the time between network access and payload deployment. Attackers can move faster than human defenders can react. 3. Cyber Resilience is the New Standard The industry has shifted its mindset from trying to achieve absolute prevention (which is impossible) to guaranteeing resilience. This shift acknowledges the persistence of ransomware. The focus is now on ensuring organizations can: Anticipate and detect threats early (low MTTD). Withstand the attack without immediate operational collapse. Recover guaranteed clean data within minutes (low MTTR). Ransomware will not disappear until the criminal model becomes unprofitable, and current data shows that attackers are highly successful and rapidly adapting their strategies.Any movement on Active Cluster with NVMe /TCP?
The last update I can find is from 2 years ago was a "maybe 2025" post. Nothing in the AC FAQ that I can find. Is this going to happen or should we just explore other conectivity options at this point? Servers are UCS with internal FC switching for now, we would like to configure uniform access. Pretty sure VMware isn't going to officially support multiple connection types for the same LUN/device/volume, unless someone knows something different.99Views0likes3Commentsrelationship between clusters and host groups
Due to circumstances beyond my control, over time and several upgrades of vSphere, there is no longer any commonality between the clusters in vSphere and what I see as host groups in the FlashArrays. So, my question is - does the API figure things out one way or another OR is my assumption that in the ideal world the clusters of hosts in VMware should match the host group names and such in Pure land? Basically some folks played shell games and since the zoning allowed it, lots of things don't line up like I once had them. I am unsure of the best way to make them sane again. thanks!208Views1like8Comments