Via its ILIO product, Atlantis Computing initially made a name for itself by slashing I/Ops ten-fold--dramatically decreasing the need for expensive SAN storage--reducing client boot times from an average of 126 seconds to 24 seconds, and de-duplicating storage, enabling the creation of persistent clients that used no more storage than their non-persistent counterparts.
Now, with the recent introduction of Atlantis ILIO Diskless VDI, the company seems within reach of the Holy Grail of VDI: the elimination of all storage-related CAPEX and OPEX, and a further decrease in desktop boot times to 12 seconds. The new product is also said to be the first solution to eliminate storage for Citrix and VMware virtual desktop images.
As the company is quick to note, with existing VDI solutions, virtual desktop images are stored on either shared SAN/NAS storage or local SAS/SSD disks, which are bogged down by limited I/Ops for VDI workloads, and tend to have brief lifespans and warranties--all of which can be deal-breakers for less affluent users.
Atlantis ILIO Diskless VDI software comes to the rescue by performing NTFS traffic processing and inline deduplication of images to run all desktops from local server memory, meaning the days of crippling disk failures are gone along with exorbitant storage requirements for rack space, power and cooling. The company claims that diskless VDI produces response times that exceed even those of the most expensive local SSD drives (MLC or SLC), while slashing VDI CAPEX to less than $200 per desktop.
Posted by Bruce Hoard on 02/01/2012 at 3:58 PM3 comments
According to the 2012-2011 Annual Salary Survey by Dice.com, on average, if you are skilled in Azure, you can make the big bucks--$102,510 per annum, to be exact. Given Microsoft's claims that there are tens of thousands of Azure installations around the world, and in light of recent rumblings about the increasingly powerful position of PaaS, this doesn't come as a huge surprise. What does come as a bit of a surprise is the Dice.com finding that Azure's reportedly red-hot cousin Hyper-V is at the bottom of the barrel, paying a paltry $81,701 to its technical practitioners.
Following Azure in respective order are Amazon S-3 ($94,843), Cloud Computing ($92,830), VMware vCloud ($88,854), Virtualization (86,669), Xen ($85,591), and VMware (82,688), and Xen ($85,591).
Only three of the categories--Virtualization, VMware, and Xen--go back to Dice's 2009-2008 survey, and among them, only Xen pays less now ($85,591) than it did in the 2009-2008 survey ($87,195). Cloud Computing goes back to 2010-2009, when it paid a relatively robust $92,663 compared to its nearly unchanged current level of $92,830. Amazon S-3, Azure, and Hyper-V, VMare vCloud are all first-timers.
What's it all mean? From what we can see, salaries are largely holding their own. What we can't see is what percentage of respondents work full-time in the categories they checked off, i.e. do these people consider themselves to be dedicated virtualization admins or cloud architects, or are they spreading their skills across multiple categories? This is especially interesting considering the movement toward making IT pros jacks of all trades as their jobs are reshaped by technologies like cloud and virtualization.
You could make the argument that Hyper-V salaries are lower because the salaries of IT pros in Microsoft shops are lower, and Microsoft IT pros are administering Hyper-V, since it comes gratis with Windows Server 2008 R2.
The Dice Salary Survey was administered online with 18,325 employed technology professionals responding between September 19 and November 21, 2011. Respondents were invited to participate in the survey through a notification on the Dice home page, and registered technology professionals were sent an email invitation. A cookie methodology was used to ensure that there was no duplication of responses between or within the various sample groups, and duplicate responses from a single email address were removed.
Posted by Bruce Hoard on 01/30/2012 at 4:55 PM0 comments
I've been wondering how things were going for cloud storage company Nasuni and its CEO Andres Rodriguez, so it was nice to see they had good financial fortune in 2011 even if they have joined Veeam among the dubious ranks of quasi-public companies that produce ersatz financial statements that substitute "bookings" for revenues.
For what it's worth, Nasuni says it increased bookings more than tenfold and doubled its number of enterprise customers. Beyond that, the company tripled the number of upgrades from existing customers and grew its international business tenfold--honest.
What went so right? Nasuni says during 2011 it changed its focus from SMBs to enterprises, and sold their enterprise storage services entirely through the IT channel, which the company says will "radically change" the traditional channel model.
Here's how it works: "Instead of making a big pop on an initial storage sale, and then making just a paltry five percent on maintenance the next year, Nasuni's partners continue to get 20-30 percent on the annual fee every year--and as their customers' storage needs grow, so do the fees and money they make."
Congrats, Andre--you're walking the walk.
Posted by Bruce Hoard on 01/25/2012 at 4:07 PM1 comments
Like most storage makers, Virsto denounces the pernicious effects of storage, which is sort of like the president of the United States saying he hates the Federal government. Of course, the demonizing part is cathartic because once the air is cleared on this matter, smart people like Virsto CEO Mark Davis can step up and offer their storage products that are designed to simplify and optimize virtual environments.
Virsto says it is the first storage software company to offer storage hypervisors, and what they have to say about them is very similar to what competitors like DataCore have to say, which is that storage hypervisors improve the utilization of hardware capacity and drive down the costs of application deployment while providing greater business agility.
Since its humble beginnings in 2007, when the company initially supported the as-yet unproven Windows Server 2008 R2 Hyper-V over VMware, the venture-backed Virsto has persevered while it carved out market share, and has now introduced Virsto for vSphere and Virsto for Virtual Servers, Hyper-V 2.0, which the company claims "allows users to improve both the efficiency and the performance of physical storage by up to 90 percent, and reduce the cost of storage in virtual environments by up to 70 percent."
Citing the over-provisioning of storage and its LUN-based -- as opposed to VM-optimized -- approach, Davis says users are frustrated and CIOs outraged that they are spending so much of their IT budgets on storage, and that is dampening enthusiasm for VDI rollouts.
Virsto claims it delivers "dramatic" performance and space savings via its new virtual storage object, the Virsto vDisk. The vDisk appears to vSphere as an "eager-zero" thick VMDK, which is the highest performance virtual storage object in the VMware environment. According to Virsto, "It delivers higher performance than native VMDKs, while only consuming storage capacity as data is actually committed, avoiding the need to pre-allocate storage capacity that often is never consumed. This approach allows IT organizations to drive storage capacity utilization up by up to 90 percent on the existing storage already in use."
The vDisk is said to deliver the benefits of thin-provisioned clones, with the production-level performance of an eager-zero thick VDMK. Virsto presents this new virtual storage object directly into the existing VM provisioning workflow in vCenter, which is transparent and fully supported by all vSphere operations.
Virsto for VSI creates a virtual storage layer that presents VHDs to virtual machines running in Hyper-V. Virsto vDisks look like native Hyper-V fixed disks, but they are thin-provisioned, high performance, cluster-aware, and support extremely scalable snapshot/clone technology -- all at the same time. "When deployed with Virsto, Hyper-V delivers higher, more predictable storage performance, and uses up to 90% less storage capacity," Virsto claims.
Other product highlights for Virsto for vSphere include seamless integration into existing VM management and provisioning workflows via VMware vCenter and VMware View, as well as native support for vSphere 4.1, and Storage quality of service tiering for optimized flexibility and cost management.
Product highlights for Virsto for Virtual Servers, Hyper-V 2.0 include native support for Hyper-V and Microsoft System Center Virtual Machine Manager in addition to seamless integration with Microsoft DPM and VSS for backup and recovery.
Posted by Bruce Hoard on 01/23/2012 at 10:25 AM0 comments
Unidesk CEO Don Bulens offers an interesting perspective on the roles of CIOs and IT pros such as DBAs and application developers, when it comes to who are the real IT movers and shakers.
His logical assessment is that because DBAs and application developers are in the day-to-day trenches, they are the power wielders. One of the things that frustrates Bulens is the focus on what CIOs say, because he believes the truth about what causes IT problems can often be found by talking to the operators of infrastructure and applications. It's why many technologies have their breakthroughs with the people who manage that particular function.
Bulens says that DBAs have driven the market adoption of products many times over, and when it comes to application developers, "Worlds have gotten turned upside down many times over by application developers embracing some new tool set, some new platform."
Who wields the power in your IT shop?
Posted by Bruce Hoard on 01/18/2012 at 9:36 AM1 comments
There are a couple of mobile virtualization developments to report. First, LG is using CES 2012 as a soapbox to demonstrate VMware virtual Android machines running on a Verizon Revolution, thus creating a personal phone that individual users manage, plus a work phone that their employers control. The word from CES is that Verizon and Telefonica will be introducing this bifurcated mobile phone sometime in the “coming months,” but at this point it is still in proof of concept stage.
The less noticed announcement comes from Red Bend Software, which is a major force in the mobile software management market. Red Bend unveiled vLogix Mobile, which it calls “the first mobile virtualization solution ready for mass-market deployment to consumer and enterprise smartphones, tablets, modems and wirelessly connected devices.”
Red Bend goes on to say vLogix Mobile is the first mobile virtualization solution that can be integrated in weeks, as opposed to months (it looks like LG and Red Bend are moving at a similar pace). The new product enables two operating systems to run simultaneously, separately and securely on the same hardware. Also, like the LG phone, vLogix Mobile plays the BYOD game by combining separate profiles for business and personal use on the same smartphone and extending support for legacy software on new platforms.
Red Bend is not yet widely known in the consumer market, but these guys have been busy pumping out more than a billion Red Bend-enabled devices, and their customers include more than 80 device manufacturers, mobile operators, semiconductor vendors and automotive companies worldwide.
Posted by Bruce Hoard on 01/10/2012 at 9:51 AM0 comments
It's hard telling just how many PaaS vendors there are out there. I found one list that includes 20, so that's a starting point. Microsoft Azure, Amazon Web Service Elastic Beanstock, Google Apps, and Force.com PaaS from Salesforce.com are all, of course, prominent players.
Currently, when People think about PaaS, they associate it with the Web development community, but Lucas Carlson, founder and CEO of AppFog has a much grander view of PaaS and its role in the cloud. In his vision, PaaS provides the last mile to the cloud, and will expand the cloud's presence by bringing many more developers to it, while promoting SaaS and IaaS opportunities.
"Clearly, to deploy large SaaS implementations, you need PaaS technology to power them," Carlson declares, adding "PaaS is the best sales tool for IaaS." He goes on to say that PaaS is more "ground-shifting" and provides greater opportunities than virtualization, and his company already has "tens of thousands of customers." All this, and the golden age of PaaS is just beginning. As he puts it, "To me, it's a Greenfield opportunity."
You'd never guess this guy is the boss of a PaaS company, would you?
For its part, AppFog recently took the wraps off an add-on program for third-party service providers that provisions the accounts purchased by developers via a single interface, and displays partner information for the customer to "easily integrate additional functionality from these third-party services into the applications they build on the AppFog platform. This will make it easier still for developers to deploy and scale web-based applications without having to become part-time IT support on the side."
Carlson believes that the PaaS company with the best ecosystem will be victorious in the market, and toward that end, AppFog has added Mongolab--which will offer its hosted MongoDB to the AppFog community--and New Relic, which says its Web application performance tool gives developers "deep, 24x7 visibility from the end-user experience all the way to a line of application code, which is a crucial capability for developers using AppFog for deploying apps to the cloud."
Posted by Bruce Hoard on 01/06/2012 at 2:51 PM0 comments
I like keeping up on users, so when this case study from Avere Systems popped up, I was intrigued because it described a straight-forward solution to a vexing problem that more than one VDI system has encountered.
It all starts with the Belchertown School district in Massachusetts, which thought it was making all the right moves. Its IT team put together a system based on five Cisco UCS systems running VMware View connected to 10 data stores hosted on a NetApp FAS2020, and supporting one-terabyte volumes, with a 20 percent snapshot reserve. Most VMDK files are held to 40GB and are typically linked clones from one of a few golden master images.
This configuration supports PowerSchool, a virtualized application utilized by teachers and students alike to access class materials, log attendance and store and back up their work and grades. All in all, a locked-and-loaded system--until the boot storm ensued when some 250 concurrent users logged on at the same time.
It typically took 15 minutes for the storm to subside. During that time, students were unable to access their class documents, teachers got errors when they tried to save their work, and the school district's small, but valiant, IT team couldn't keep up .
They evaluated bigger, more updated NetApp filers, but the costs were prohibitive, and the time required to implement them was unacceptable. The remedy was to be found at a VMware user group meeting in Maine, where they discovered Avere's NAS Optimization solutions.
It was a Eureka moment for Scott Karen, the school district's director of Technology. As he puts it, "Avere was able to get us an FXT Series node for evaluation almost immediately. We tested it, and found it eliminated the boot storm and turned 15 minutes of logging in per class into about three minutes with no more write errors or corrupted files. Overall latency is no longer an issue. And it didn't require me to rethink and re-engineer my storage network."
Karen moved quickly, and within three weeks of choosing Avere, the FXT 2500 two-node cluster was up and running. The nodes are configured in read/write mode during the school day, and at 6:00 P.M., the cluster automatically switches modes to read-only in an effort to support the district's existing backup strategy with Veeam. Twelve hours later, the cluster switches back to read/write mode for maximum performance during the school day.
The FXT 2500 clusters are reliable, they don't need tweeking, and Karen hasn't had to touch them since they were implemented. He refers to them as "magic boxes," adding "Because of Avere's tiered storage architecture, all future disk purchases can be lower cost SATA drivers rather than higher cost SAS, so Avere has not only solved our performance problems, but will be saving us money going forward."
It is, as they say, all good.
Posted by Bruce Hoard on 01/04/2012 at 3:43 PM2 comments
The Virtual Geek is an EMC blog written by a funny guy who is coincidentally known as the Virtual Geek.
His Official Unofficial VMware Storage Survey Results came out last week, and there are, as they say, some interesting data points here that dovetail with my "vSphere 5: Slow to Roll? blog, which also posted last week. In it, I asked the question, "Could it be that we are underestimating the rigors of implementing vSphere 5?"
Virtual Geek answered that question in an indirect fashion when he revealed that 59.2 percent of the 1,935 respondents said they were already using vSphere 5. That's a lot of users who are implementing vSphere 5 post haste, since it only GA'ed in late August. Working backwards, 85 percent of respondents identified themselves as vSphere 4.1 users, followed by 24.3 percent saying they were vSphere 4.0 users, 12 percent saying they were on V13.5, and 1.4 percent reporting they were pre-V13.5 users.
Methodology nugget: The number of virtual machines at respondent sites ranged from four to 200,000.
Methodology non-nugget: Virtual Geek did not describe respondents in a meaningful fashion, other than saying they came from 65 unique countries, and 941 unique cities.
In response to the question, "Do you use any other Virtualization Technologies?", 48.8 percent cited the rapidly up-and-coming Hyper-V, followed by 26.5 percent noting XenServer, which, it would seem, is benefiting handsomely from its close relationship with VDI meister XenDesktop.
Virtual Geek's survey also revealed that a lot of the low-hanging virtualization fruit has indeed been picked. To wit, when asked what percentage of their x86 environments were virtualized, 11.3 percent said 100 percent, 32.2 percent said 90 percent, 80 percent replied 23.7 percent, 60 percent noted 18.3 percent, 40 percent said 8.3 percent, 20 percent said 4.2 percent, and 10 percent reported that two percent of their x86 environments were virtualized.
With a nod toward Virtual Geek, I point out that EMC topped the list of replies when respondents were asked to check off their favorite vendor of choice in virtualized server environments. EMC basically lapped the field with 41.1 percent of the vote, followed by the usual storage suspects, NetApp, HP, Dell, IBM and Hitachi -- none of whom received even half the responses chocked up by EMC. It's good to be King.
Posted by Bruce Hoard on 12/20/2011 at 7:21 PM1 comments
Now that Veeam has announced that its Backup & Replication v6 product provides native support for Hyper-V, the company is looking to ingratiate itself with IT pros who have Microsoft certifications, and expand sales by offering them free software licenses for demo use in home labs. Veeam did the same thing last year for VMware-certified pros.
Overall, these free, not-for-resale, two-socket software licenses are now available to VMware vExperts, VMware Certified Professionals, VMware Certified Instructors, VMUG members, Microsoft Most Valuable Professionals, and Microsoft Certified Professionals.
Veeam has been extremely bullish on Backup & Replication dating back to its pre-production days, and says it was downloaded over 15,000 times during its first week of availability.
In a canned quote, Microsoft MVP Derek Schauland seemed jubilant, saying "Veeam for the Microsoft community? Awesome! Having a program giving NFR licenses of Veeam software to the Microsoft community will be invaluable to the home lab and test needs if IT pros in pursuit of training, blogging and other activities."
The licenses are available here.
Posted by Bruce Hoard on 12/15/2011 at 10:24 AM3 comments
I realize that I recently blogged on the results of a survey from Nasuni about cloud-based data security, but I thought readers would be interested in the company's latest research findings, which are based on the results of an ongoing 26-month stress test of 16 major cloud storage providers (CSP).
According to Nasuni--which uses raw cloud storage from these companies--only six of the 16 CSPs passed the test, meaning they provided the "minimum level of performance, stability, availability and scalability that organizations need to take advantage of the cloud for primary storage, data protection and disaster recovery." The six with passing grades are Amazon S3, AT&T Synaptic Storage as a Service (powered by EMC Atmos), Microsoft Windows Azure, Nirvanix, Peer 1 Hosting (also powered by EMC Atmos), and Rackspace Cloud.
The two top performers are Amazon S3, which ranked highest across the board, and Azure.
Nasuni will not reveal the names of the 10 CSPs that flunked out.
Regarding methodology, the tiered tests are designed such that each CSP has to pass an initial test before moving on to the next one. The five testing stages are:
- API integration, which ensures it is possible to test the service
- Unit testing, wherein larger software components are broken down into their building blocks--units--and then tested for inputs, outputs and error cases
- Performance testing, which measures the speed of cloud interaction, meaning how rapidly data moves back and forth to the cloud, and the reaction to high stress levels
- Stability testing, which measures the prolonged reliability of CSPs
- Scalability testing, which reveals how the CSPs react to high object counts
"Though Nirvanix was 17% faster than Amazon S3 for reading large files, and Microsoft Azure was 12% faster when it comes to writing files, no other vendor posted the kind of consistently fast service across all file types as did Amazon S3," Nasuni said.
In addition, S3 had the lowest number of outages and the best uptime, and was the only tested company to register a 0.0% error rate in both writing and reading objects during scalability testing. Nasuni also said that while Azure has a slightly faster ping time than S3--which Nasuni attributes to the fact that S3 is much more heavily used than Azure--S3 still maintained the lowest variability.
Just as an aside, I find it interesting that in its current company description appearing at the bottom of its press releases, Nasuni fails to mention the word "cloud" once, while just over a year ago in early October, it started out a new product announcement for the company's Nasuni Filer 2.0 by calling itself the "creator of the storage industry's leading cloud gateway." Nowadays, Nasuni refers to itself as "a next generation enterprise storage company."
More information on the Nasuni Stress Tests is available at www.nasuni.com/cloudreport.
Posted by Bruce Hoard on 12/13/2011 at 4:17 PM0 comments
Could it be that we are underestimating the rigors of implementing vSphere 5? I ask that because during a briefing I had with vKernel this week, when I asked Product Marketing Manager Alex Rosemblat what he was hearing about vSphere 5 migrations, he replied, "We have very few customers who are using it extensively yet." This, from a guy whose company sells heavily into dedicated VMware environments.
I'm not implying that there are serious hidden problems here. After all, vSphere 5 officially hit the streets in late August, only some three months ago, but it just seems a little odd that Rosemblat hasn't felt a stronger product pulse for the latest and greatest iteration of VMware's primo virtualization environment. He did say that some users were doing pilots, so maybe IT organizations are making sure they've got vSphere 5 down pat via a proof-of-concept approach before they mount a serious migration.
Here at Virtualization Review, we have been doing our best to keep readers abreast of this hot topic via a new, dedicated vSphere 5 page, and blogs from our How-to Guy, David Davis, and Virtual Insider Eli Khnaser. David most recently wrote a blog entitled, "How to Monitor vSphere 5 vRAM Pools," while Eli produced "ESX-to-ESXi Migration Key to Success? The ESX System Analyzer."
As vSphere 5 gains momentum, we will continue to report the news and analyze the issues on this pressing topic.
Posted by Bruce Hoard on 12/08/2011 at 6:41 PM4 comments
iWave Software offers an alternative to users who want to develop their own cloud-based storage services as opposed to hiring a service provider for the task. In addition to giving companies the comfort of knowing they have architected their own, on-demand, cloud storage solutions, iWave Storage Director 1.5 adheres to the immutable rule that states "Thou shalt not displace legacy storage infrastructures."
Storage Director 1.5 caters to budget-sensitive customers by automating the tasks associated with provisioning, reclamation and remediation for storage. In addition, it enables storage admins to push the self-service button, which enables them to automate storage services in private storage cloud environments that are available to storage consumers. In this way, admins are able to cost-efficiently leverage unified workflows across vendor products and within organizationally defined policies for regulatory compliance. Bottom line: Fewer admins are required, and those who remain can focus on other, more core-competency-related tasks.
Reduced storage outages are another benefit provided by Storage Director 1.5, because it ensures best practices are automatically followed with each request for provisioning. This eliminates configuration errors during provisioning. Throw in finding and reclaiming unused storage, and improving end-user satisfaction by reducing the time required for provisioning new storage from weeks to hours, and you have a full solution set.
Company quote: "Our solution gives storage administrators the ability to develop a fully automated storage services catalog based on best practices, making storage easy for users to consume and for administrators to manage."
Posted by Bruce Hoard on 12/06/2011 at 3:53 PM0 comments
Gone are the days of DR tests being done every--say, three months, no make that every six months, no maybe once a year--but it costs so darned much and takes so long, let's just blow it off. Well, those days aren't gone for everybody. In fact, as far as we can tell, they are only gone for customers of VirtualSharp, which assessed the sorry state of DR testing and came up with a product--ReliableDR 3.0 in its latest incarnation--that handles the entire test process, end to end, automatically. No muss, no fuss.
After hearing how one of VirtualSharp's Fortune 20 insurance customers developed an ongoing, automatic DR test that runs every three hours eight times a day--that is smokin'--I'm inclined to believe Virtual Sharp CEO Carlos Escapa when he refers to his agentless product as a "a paradigm shift altogether."
More good news: all that productivity comes bundled with savings of somewhere in the neighborhood of $50,000 to $100,000 for that prescient, F20 customer, who can now stay home and pull for the Green Bay Packers rather than worrying about DR tests that previously required a team of application owners to up and and head off to a Sungard DR center for a gala 3-4-day weekend of humdrum walkthroughs.
Scarpa says his agentless product "delivers an integrated, virtual, disaster recovery solution across the IT stack, from storage to application, that can be delivered for cloud environments using a zero-footprint, self-service architecture. It's good to be king.
ReliableDR 3.0 improves on its predecessor by adding multenancy, a Web Oriented Architecture, Application-specific SLAs, role-based access control and embedded dashboard (probably from a Mercedes).
The product will be available during Q1 of 2012 directly from VirtualSharp and its authorized channel partners. Existing 2.x customers will be entitled to a free upgrade to 3.0 based on their current maintenance agreements.
Question: Would your company benefit from automated DR testing?
Posted by Bruce Hoard on 12/01/2011 at 5:46 PM0 comments
Cloud storage provider Nasuni is fighting the good fight when it comes to selling cloud storage. As everybody knows, a lot of users are still hesitating to move their data offsite and onto the cloud, but it does happen, or CloudSwitch would have not been acquired by Verizon.
In a recent study conducted by Nasuni of 451 IT decision makers at North American enterprises, respondents echoed the concerns that many companies have stressed, e.g. 81 percent of respondents are concerned about data security in the cloud, while 48 percent were concerned about the level of control over cloud-based data storage. Going on, the survey revealed that only 43 percent of respondents have plans to store their data in the cloud over the next year.
So how is this dire news good for Nasuni? The good news is that according to Product Evangelist Chris Glew, Nasuni is ready to ride to the rescue if overriding concerns about cloud storage force users into buying consumer storage solutions that are less secure than the enterprise-grade cloud storage solutions offered by Nasuni.
"Our recent survey makes painfully clear the quandary the cloud presents to IT leaders," Glew opines. "They clearly understand the promise of cloud storage for cost savings, off-site backup, unlimited scale, simpler IT management, and on-demand provisioning, but they are also rightfully concerned about the security of their data and whether they have control over it at all times."
Which takes us back to the longstanding tendency of rogue users who like the idea of a technology--such as cloud storage--but opt for lower quality consumer options without the permission of IT. By "consumer," Glew means services that are typically used to share information among friends--and are potentially disastrous for businesses.
"The survey demonstrates the need for a new, cost-effective approach to storage that harnesses the unlimited capacity of the cloud in a way that guarantees data is always secure, always manageable, and always under IT's control," says Glew.
Nasuni is only too happy to help users save themselves from their penurious ways.
Posted by Bruce Hoard on 11/29/2011 at 3:30 PM0 comments