Slashdot Mirror


User: Bengie

Bengie's activity in the archive.

Stories
0
Comments
6,462
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 6,462

  1. Re:What ever happened to training? on IT Positions Some of the Toughest Jobs To Fill In US · · Score: 1

    The average worker at my job is 8 years and the company is only 15 years old and exploded in growth back in 2000. Holding on to workers is easy. Treat them with respect, work as a team, and pay them fairly. Nearly all of our turn-over is caused by retirement, then there's the Google/Microsoft poachers, but we just wish those ex-employees luck.

    It's common to hear our customers tell us we're the best company that they've worked with. Everyone is friendly and willing to help, even if it's not their position. With strong employee retention, there is a strong social network, so the average person knows whom to contact to figure out a problem.

    When one part of the company needs help, they get other departments offering to help all the time. Even managers and executives get their hands dirty if they have the ability to help.

    Working in this kind of environment is worth A LOT of income to me.

  2. Re:Salaries on IT Positions Some of the Toughest Jobs To Fill In US · · Score: 1

    truth in both yours and gp

  3. Re:The Destruction of the Sky on Solar Geoengineering Could Lead To Whiter, Brighter Skies · · Score: 2

    Astronomers everywhere hate this idea.

  4. Re:Salaries on IT Positions Some of the Toughest Jobs To Fill In US · · Score: 1

    DailyWTF material right there

  5. Re:Salaries on IT Positions Some of the Toughest Jobs To Fill In US · · Score: 1

    When I worked in IT, the other depts bought us pizza or baked us cookies.. nom nom nom. They were always so happy when we fixed stuff for them. A well fed IT is a happy IT. When IT isn't happy, the world crashes down around you.

  6. Re:How DARE they! on The Poor Waste More Time On Digital Entertainment · · Score: 1

    There's a difference between realistic thinking, and idealistic thinking

    Could also be said as.. "If we ignore human nature". Many people are willing to not care about "strangers". It's the "I will run you over to save a penny" mentality that hurts everyone.
    We need some f'n real life carebears to fix the state of humanity.

  7. Re:Has the available hardware caught up yet? on German Cable ISP First To Deliver 4700Mbps Internet Connection · · Score: 1

    Saying an off-the-shelf router can handle 1Gb/s is like saying a smart car can handle 180mph because the speedometer goes that high. Just because they have 1Gb interfaces does not mean the low grade cpu can process and route packets that fast.

  8. Re:I'm so glad on German Cable ISP First To Deliver 4700Mbps Internet Connection · · Score: 1

    Go look at my post http://news.slashdot.org/comments.pl?sid=2886193&cid=40168401
    The data on my NIC, according to Intel's documentation, is that it takes up about 1mm^2 of the 45nm based chip. I don't know how many transistor you can pack into 1mm^2 on a 45nm process, but I don't expect it to be too complex. The whole-sale value of my NIC was listed about $5. You need to purchase some decent NICs.

  9. Re:I'm so glad on German Cable ISP First To Deliver 4700Mbps Internet Connection · · Score: 1

    PCIe x1 has the same throughput as PCI32/33

    PCIE 1.0 1x-lane is 250MB/s up and down and is dedicated between the chipset and device. Peak bandwidth ~500MB with very little overhead
    PCI32/33 is 133MB/s up or down and has a shared interrupt style TDMA to the chipset. Peak bandwidth ~133MB minus a decent amount of overhead

  10. Re:I'm so glad on German Cable ISP First To Deliver 4700Mbps Internet Connection · · Score: 1

    Anecdote: My PC is 5 years old, cost around $600 from Best Buy, has an integrated Intel NIC, and IPerf can do about 1.5Gb/s(up+down at the same time) for ~10% cpu(almost a full core). You should see SMB2.0 transfer a sustained 116MB/s at 1.5% cpu load from my harddrive.

  11. Re:Modest? 100Mbit? on German Cable ISP First To Deliver 4700Mbps Internet Connection · · Score: 1
    Most rust-bucket harddrives can handle ~100MB/s which is ~800Mb/s, which is quite a bit faster than 30Mb. No striping required. My SSD can handle ~550MB/s which is 4400Mb/s, which is almost the entire 4700Mb/s connection that they have. now finding useful data that you can push at those speeds is quite hard.

    When I first got a 10Mbps connection I couldn't use it to the max because my old computer couldn't handle it

    Must have been really old. Not only can I peg my 100Mb internet connection with little effort of Steam/Blizzard/Cloud/etc, but I can easily peg my 1Gb ethernet by grabbing a single file to SMB copy on my 5 y/o computer.

  12. Re:Wow, I'm amazed... not. on German Cable ISP First To Deliver 4700Mbps Internet Connection · · Score: 3, Interesting

    subscribers' bandwith is often shared

    All ISPs are shared at some level. I'm assuming you mean the node bandwidth is shared/over-subscribed?

    They used 12 modems and thus 12 seperate channels

    Many modems can bond up to 8 virtual channels on the down stream. As far as we know, it could have been anywhere from 1-8 8mhz channels per modem. Even with a single 8mhz channel, DOCSIS3 can bond 8 virtual CDMA channels for a combined bandwidth of 8X50Mb/s=~400Mb/s (EuroDocsis). DOCSIS3 has no limitation on how many channels may be bonded, but I'm not sure of any modem that supports more than 8 right now.

    Still cheaper/simpler to use fiber, but cable can manage some crazy high speeds if you throw enough tech/money at it.

  13. Re:How DARE they! on The Poor Waste More Time On Digital Entertainment · · Score: 1

    I mostly agree. I do what is fun and it just so happens what I'm good at and enjoy pays well.

  14. Signing away on Windows 8: More EULA, Fewer Rights. · · Score: 1

    Signing away your right to a class-action suit should only be limited to settlements, not as entry agreements.

  15. Re:Oh, joy. on Hundreds of IP Addresses Make Pirate Bay a Hard Target · · Score: 1

    Copying is a human/natural right. How do you think ANYTHING has come to be? Language, math, social interaction, life itself.... Copying.

    To be anti-copying is to be anti-life.

    No, copying someone's answers to a test doesn't count. The purpose of a test is to measure a person, not their neighbor. "cheating" via copying invalidates the measurement which is a waste of other people's time and is akin to lying.

  16. Re:I wonder if they have IPV6 support on Hundreds of IP Addresses Make Pirate Bay a Hard Target · · Score: 1

    The company proxy wouldn't know the difference between a high-end verified cert and a cheap anonymously purchased cert. Might as well just allow self-signed certs.

    Since we're going there. If the company is going to allow self-signed certs and not show the cert to the end user, why even use SSL?

    The logical conclusion.

  17. Re:Exactly why we don't need IPv6 on Sales of Unused IPv4 Addresses Gaining Steam · · Score: 1

    addressing != processing
    While your post is correct, it's off-topic.

  18. Re:Exactly why we don't need IPv6 on Sales of Unused IPv4 Addresses Gaining Steam · · Score: 1
    Back in 2008 there were tests by AARNet checking that multi-cast was working across the public internet and over oceanic links.

    Thus each user of an IPv6 subnet automatically has available a set of globally routable source-specific multicast

  19. Re:open standards for hardware on Why Open Compute Is a Win For Rackspace · · Score: 1

    I would love to have a 12v server. UPS is just an individual battery and/or a battery bank.

  20. Why 128bit? on Sales of Unused IPv4 Addresses Gaining Steam · · Score: 1

    The goal with IPv6 allocations is to give more than enough IPv6 address space. They want to prevent an organization from needing to come back to request more and also to avoid fragmented route prefix advertising in the core Internet routing tables.

    Here your answer why 128bit addresses are used. You don't hand out address based on need, you give something so large that no one could use it all.

  21. Re:Hoarding? on Sales of Unused IPv4 Addresses Gaining Steam · · Score: 1

    Exactly. They were issued these blocks in good faith. They are hoarding them, and they dont own them - they need to return them so they can actually be used

    You make it sound like they are doing something they shouldn't. In reality, IPv4 was a proof of concept and anyone was allowed to join the "beta" network. "Here have some IP addresses, free of charge, they're yours until we make the "real" protocol.

    IPv4 takes off and now these companies who got these large /8 blocks have no restrictions because that was the contract at the time. We can't just change the contract on them because we don't like it anymore.

  22. Re:IPv4 forever? on Sales of Unused IPv4 Addresses Gaining Steam · · Score: 1

    "Hey guys, we have a new toy, here's some numbers you can have". 25 years later.. "shiat"

    It may have been /., but some site had an interview with the creator of IPv4 and he said it was just a proof-of-concept, but it took off before IPv6 was finalized. Once IPv4 become too popular, IPv6 went to the back-burner for a bit.

  23. Re:Exactly why we don't need IPv6 on Sales of Unused IPv4 Addresses Gaining Steam · · Score: 1

    Can you do so without changing IP addresses? IP renumber is very expensive for network/server admins and it is a very common thing with merging corp networks.

  24. Re:Regulation on Sales of Unused IPv4 Addresses Gaining Steam · · Score: 1

    Sorry, our version of the internet no longer works. Sounds like a great beginning to a class-action suit after many software/consoles/etc break. I'm sure Ipv6 will take off before it becomes much of an issue. Anyway, many network engineers have talked about how ISP level NAT for broadband connections is more expensive and more of a head-ache than just switching to IPv6.

  25. Re:sounds a bit facebooky on Sales of Unused IPv4 Addresses Gaining Steam · · Score: 1

    no more than a NAT can be tracked.