Dell Inspiron M5010 AMD 2.1GHz, 3GB, 320GB

 

For more detailed configuration and pricing feel free to  Contact Vendor

Two billion-transistor beasts: POWER7 and Niagara 3

     By Jon Stokes |

A 300mm Power 7 processor wafer

In years past, an ISSCC presentation on a new processor would consist of detailed discussion of the chip's microarchitecture (pipeline, instruction fetch and decode, execution units, etc.), along with at least one shot of a floorplan that marked out the location of major functional blocks (the decoder, the floating-point unit, the load-store unit, etc.). This year's ISSCC is well into the many-core era, though, and with single-chip core counts ranging from six to 16, the only elements you're likely to see in a floorplan like the two below are cores, interfaces, and switches. Most of the discussion focuses on power-related arcana, but most folks are interested in the chips themselves.

In this short article, I'll walk you through the floorplan of two chips with similar transistor counts—the Sun's Niagara 3 and IBM's POWER7. Most CPU geeks will already know a lot of the information I'll give below, but many readers will appreciate having it all together in one place.

Niagara 3: threads and I/O

Sun's Niagara 3

Sun's 1 billion-transistor, 16-core Niagara 3 processor is a great example of modern multiprocessor-turned-SoC (system on a chip). Everything about this design is focused on pushing large numbers of parallel instruction streams and data streams through the processor socket at once. The shared cache is small, the shared pipes are wide, and the end result is a chip that's all about maintaining a high rate of flow, and not one that's aimed at collecting a large pile of data and chipping away at it with heavy equipment.

Each of the 16 individual SPARC cores that make up Niagara 3 support up to eight simultaneous threads of execution, for a total of 128 threads per chip. Logically, the chip is laid out so that all of the cores communicate with a unified 6MB L2 cache via a crossbar switch that's placed in the middle of the chip. This combination of cores and L2 connected via a switch forms the basic compute architecture of the SoC.

So that the chip can talk to the outside world, the L2 caches are connected to a variety of I/O interfaces: memory, PCIe, 1G/10G Ethernet, and coherency links. All told, those links can push a total of 2.4Tb/s worth of data through a single Niagara 3 socket—that's a lot of bandwidth, but you need it to feed that many threads. Let's take a quick look at each of these I/O links in turn.

Coherence: Niagra 3's coherence links are the equivalent of the QuickPath Interconnect (QPI) on Intel's Nehalem parts, or of HyperTransport for AMD. These links can be used to connect up to four of the chips together without any additional routing chips (this is what's meant by saying Niagara 3 can be used in a four-socket glueless configuration). Each Niagara 3 chip has two 1.6GHz coherence controllers, which are connected to six coherence links. Each individual link consists of 14 unidirectional lanes that give the link a total bandwidth of 9.6Gb/s.

Memory: Also attached to the L2 is are two DDR3 memory controllers, each of which hosts two memory channels, for a total of four channels of DDR3.

PCIe and Ethernet: A PCIe controller supports two 5Gb/s PCIe ports, and an Ethernet controller supports two 1G/10G Ethernet ports.

IBM's POWER7

IBM's POWER7

At 1.2 billion transistors, IBM's new 45nm POWER7 processor is only a little bigger than Niagara 3, but it couldn't be more different. If Niagara 3 is an army of guys with shovels, POWER7 is a giant bulldozer.

POWER7 has only half the cores (eight) and one quarter of the threads (32) of Sun's chip, but that doesn't mean it falls short in the horsepower department. Each POWER7 core has a ton of very fast execution hardware, and the overall layout of the machine's very wide execution core is a straightforward evolution of the design that I first described in a series of articles on the PowerPC 970. (I talked more about POWER7's execution core in an earlier article comparing it to Intel's Tukwila Itanium.)

Where Niagra 3 keeps a large number of relatively weak cores busy by moving data onto and off of the chip using ample I/O resources, POWER7's approach to feeding a smaller number of much more robust cores is to cache large amounts of data on-chip so that the cores can grind through it in batches. This being the case, POWER7 has the most remarkable on-chip cache hardware of any processor on the market.

First in the chain is the 32KB L1 data cache, which has seen its latency cut in half, from four cycles in the POWER6 to two cycles in POWER7. Then there's the 256KB L2, the latency of which has dropped from 26 cycles in POWER6 to eight cycles in POWER7—that's quite a reduction, and will help greatly to mitigate the impact of the shared L3's increased latency.

The POWER7's L3 is its most unique feature, and, at 32MB, it's positively gigantic. IBM was able to cram such a large L3 onto the chip by making it out of embedded DRAM (eDRAM) instead of the usual SRAM. This decision cost the cache a few cycles of latency, but in exchange IBM got a 3.5x improvement in power efficiency and a 3x improvement in cache density. IBM has actually been talking up the use of eDRAM for on-chip cache since at least 2002, so in this regard POWER7 represents the fruition of years of work on this approach.

On the I/O side, POWER7 features two DDR3 memory controllers that can do up to 100GB/s total. The chip's SMP links (the same as Niagara's coherence links) can do 360GB/s (or almost 2.9Tb/s) of bandwidth, but this amount appears to be divided between internal and external SMP links. The chip doesn't contain the other I/O options—PCIe or Ethernet—that Niagara has.

Ultimately, these two server-class processors show that there are two very different ways to spend a billion transistors, and each design will be good for different applications. Sun's Niagara is aimed at networked server operations where lots of simultaneous, lightweight requests have to be serviced—databases, Web servers, and the like. In contrast, POWER7 has the horsepower to grind through a smaller number of more compute-intensive tasks at a high rate of speed. Both parts have their place in the server ecosystem of 2010.

 

 

The New Segregation Debate

 

Can educating girls and boys separately fix our public schools, or does it reinforce outmoded gender stereotypes?

(Page 1 of 3)

Peter M. Fisher / Corbis

If you thought charter schools and ending teacher tenure were controversial fixes for the American school system, see what happens you bring up the idea of educating boys and girls separately. With male academic achievement declining by almost every measure, and their scores possibly dragging down national averages, administrators are taking a fresh look at same-sex classrooms and the concept that boys and girls might do better when they’re apart. Why is it such a hot-button topic? Well, because it goes against 30 years of thinking, and smacks of “separate but equal” education.

The advocates of the single-sex approach are surprising, as are the foes. Among many liberal thinkers, gender segregation sounds like regressing to a time when girls were educated in finishing schools and had access to neither the number, nor caliber of schools available to boys. Plus, the notion that boys and girls learn differently—touted by some as the primary rational for gender separation—goes against one of feminism’s (at least the 1970s version) main messages. To say that there is something inherently different between boys and girls is, for many, tantamount to saying that women are the weaker sex.

For these reasons, Democratic politicians spent decades fighting vehemently against loosening legislation to allow public schools to offer same-sex classes. But in 2001, Sen. Hillary Clinton linked the issue to class—citing an unfairness in the fact that single-sex education is available as a choice only to those who can afford private-school tuition. Clinton, a graduate of all-women’s Wellesley College, joined forces with Republican Sen. Kay Bailey Hutchison to successfully bring about legislative change. Since then, the number of public schools offering same-sex classes has grown from 11 to 540—still a relatively small figure in the big picture, but a jump of more than 4,000 percent nonetheless.

Related: America's Best High Schools 2010 »

This number would be even bigger were it not for the ACLU, which has successfully convinced dozens of districts not to adopt single-gender classrooms. “Our concern is that once you separate boys and girls you are telling them that there is some inherent difference such that they need to be educated separately,” says Lenora Lapidus, head of the women’s-rights arm, which is spearheading the investigation. “When public schools do this, it’s the government reinforcing gender stereotypes.” Lapidus’s division also has open cases against districts they believe are violating the admittedly loose terms of the ammended legislation—in those cases, their beef is that single-sex classes are being forced on children, and parents aren’t being given a legitimate coed option. But Lapidus says that she believes public schools shouldn’t offer single-sex classes under any circumstances—whether it’s a choice or not.

The ACLU’s opposition perplexes advocates of separate classrooms. “The ACLU has become increasingly deranged over the years,” says Leonard Sax, the head of the National Association for Single Sex Public Education. “And by deranged I mean out of touch with reality.” Sax also can’t understand why the National Organization of Women (NOW), which advocates choice in reproductive rights, would be against giving parents more options when it comes to education. “We are the pro-choice movement in this debate—we don’t believe that every child should be in same-sex classrooms, but every parent should have a choice.”

But what are parents choosing exactly? In some cases they’re getting not just separate rooms for girls and boys, they’re getting a modified curriculum and even classroom structure based on what proponents see as gender differences in learning. And that’s where things get thorny. According to Sax, some of the most successful all-boys classrooms have been those in which boys are allowed to move around. He cites one classroom that uses desks that can be raised or lowered, so boys can lie on the floor, sit, or stand up while they work. Girls’ classrooms, meanwhile can encourage quiet study, which some say is intrinsic to female behavior and caters to girls’ strengths. The question is whether this is catering to strengths or encouraging old-school stereotypes.

Sax and others will hasten to add that while the crisis among boys may have gotten the most media attention in recent years, there continue to be areas in which girls lag behind, particularly in math and science. In computer science, the divide has increased in the last few decades—only 18 percent of those taking the AP computer science test are female, down from 34 percent in 1987. According to Sax, separating boys and girls in those subject areas can be beneficial to both genders, and that’s, in part, the source of his frustration with the ACLU. “If the single-sex format offends her delicate political sensibilities, what would she recommend?” he says of Lapidus’s position. “Or is she content that boys outnumber girls four to one in this country in the sciences?”

Whether or not it works is a contentious issue—those in favor cite research supporting it, those opposed dismiss that evidence as junk science. Pedro Noguera, a professor at NYU’s Steinhardt School, recently completed a study on same-sex education for boys of color and found no evidence that separating them was better. Noguera is critical of advocates like Sax, saying that their theories are “far-fetched” since they rely more on observations of student behavior, than hard science proving that childrens’ brains work differently depending on their gender. “The schools that developed for girls had a clear theory: girls needed to be empowered, needed to express themselves,” he says. “There’s no theory about why [same-sex education] is good for boys … It’s a lot of pseudoscience that is not rooted in anything.”

But Sax and others cite some compelling statistics: in one three-year pilot project in Florida, boys in a coed class scored 37 percent on the state standardized test, while those in a boys-only class scored 85 percent. The girls’ scores jumped too: from 59 percent to 75 percent. And teachers, by and large, seem to love it. They say they spend less time on discipline, and are better able to engage their students. Lapidus argues that such studies can be attributed to other factors, like smaller classrooms and better teacher training, and that teachers who advocate for gender-segregation are just “parroting back” what they’ve learned from Sax and others.

Regardless of the mixed research, the interest in single-sex classrooms shows just how desperate teachers and administrators are to find a cure to the oft-lamented "problem with boys." By just about every metric, boys are, and have been for perhaps a decade, lagging tremendously behind girls in terms of academic achievement. They consistently score lower GPAs, college-admissions rates, and fare worse in reading and writing. And it’s not just a problem for them; their scores aren’t helping the country’s plummeting academic ranking as compared to the rest of the developed world.

The gender gap goes far beyond high school. Today, women make up nearly 60 percent of college students and they’re much more likely to go on to pursue advanced degrees. In fact, higher education has become so tipped in favor of females that in November, the U.S. Civil Rights Commission launched a probe into whether colleges and universities were discriminating against women in favor of men in order to even out the balance.

Some blame female success itself for boys’ lagging achievement, arguing that girls have gotten ahead at the expense of boys. As the American Enterprise Institute’s Christina Hoff Sommers recently put it, “feminists are constantly trying to knock down doors that are already open. And it’s young men who pay the price.” Others, of course, argue against this kind of zero-sum thinking. Where boys might be falling behind by academic metrics, girls are more likely to report high stress levels, eating disorders, and behaviors like cutting. “Boys have problems, girls have problems,” Sax says. “Both are disadvantaged, but they’re disadvantaged in different ways.”

After seeing the same gender issues in Milford, Ohio, a suburb of Cincinnati, Gary Knell began looking into possible solutions for his school district—which has 6,200 students enrolled in kindergarten through high school. Last year, Knell organized a “gender survey” of students, teachers, and community members, measuring their feelings about how to treat the issue. Seventy percent of the 232 parents who responded agreed that gender differences should be addressed at the schools, but less than a quarter said that the district should consider single-sex classrooms. In contrast, 34 percent of the 232 middle-school students who responded said they would participate more and at a higher level in a single-sex classroom. That parents are less amenable than their kids isn’t entirely surprising. “Whenever we talk about differences, there’s a politically correct mode that differences mean someone is inferior and someone superior,” Knell says. “[But] boys and girls are different. We should recognize that and modify our teaching approaches.”

Ultimately, figuring out what method is best for boys and girls may mean identifying and tearing down gender stereotypes where they begin—with the kids themselves. And figuring out what a boy thinks it means to be a man, and what a girl thinks it means to be a woman, are questions that are probably too big for most public school to take on. But even if some adults are hesitant on the topic, the kids aren’t afraid to speak bluntly. “Boys want attention,” a sixth-grade boy in Milford says. “And they don’t really want an education. They want a cool job when they grow up. Like being a rock star or a basketball player.” His female classmate agrees. “For guys I think that if you’re really smart and you get good grades, it’ll be like, ‘Oh you’re a nerd!’ But girls, you’ll get like a pat on the back if you get all As. I dunno, guys just don’t think it’s that cool.” Therein lies a problem for which there may be no quick fix.

Find the author on the Web at The Equality Myth.

The New Segregation Debate - Newsweek